ceph/doc/changelog/v0.94.6.txt

4028 lines
137 KiB
Plaintext

commit e832001feaf8c176593e0325c8298e3f16dfb403 (tag: refs/tags/v0.94.6, refs/remotes/gh/hammer)
Author: Jenkins Build Slave User <jenkins-build@jenkins-slave-wheezy.localdomain>
Date: Mon Feb 22 21:10:17 2016 +0000
0.94.6
commit 7abb6ae8f3cba67009bd022aaeee0a87cdfc6477
Merge: ceb6fcc a8fc6a9
Author: Loic Dachary <loic@dachary.org>
Date: Mon Feb 8 14:19:57 2016 +0700
Merge pull request #7501 from SUSE/wip-14624-hammer
hammer: fsx failed to compile
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit ceb6fcc5f2828995bfe59bfc4e206010a4fe3230
Merge: 2f4e1d1 f2ca42b
Author: Sage Weil <sage@redhat.com>
Date: Fri Feb 5 21:10:46 2016 -0500
Merge pull request #7524 from ktdreyer/wip-14637-hammer-man-radosgw-admin-orphans
hammer: doc: regenerate man pages, add orphans commands to radosgw-admin(8)
commit 2f4e1d1ff8e91fc2ee9c23d5a17c3174d15a7103
Merge: 2ca3c3e 9ab5fd9
Author: Orit Wasserman <owasserm@redhat.com>
Date: Fri Feb 5 10:30:22 2016 +0100
Merge pull request #7526 from ceph/wip-14516-hammer
hammer: rgw-admin: document orphans commands in usage
commit 2ca3c3e5683ef97902d0969e49980d69c81b4034
Merge: 02353f6 5c8d1d7
Author: Loic Dachary <loic@dachary.org>
Date: Fri Feb 5 12:47:33 2016 +0700
Merge pull request #7441 from odivlad/backport-pr-14569
[backport] hammer: rgw: Make RGW_MAX_PUT_SIZE configurable
Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com>
commit 02353f6baa460949d7dd2738346d8d748401bc5b
Merge: f3bab8c 0e1378e
Author: Loic Dachary <loic@dachary.org>
Date: Fri Feb 5 12:46:54 2016 +0700
Merge pull request #7442 from odivlad/backport-pr-14570
[backport] rgw: fix wrong etag calculation during POST on S3 bucket.
Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com>
commit 9ab5fd9d67da43e986489e4f580a597dd8cb551e
Author: Yehuda Sadeh <yehuda@redhat.com>
Date: Mon Feb 1 16:33:55 2016 -0800
rgw-admin: document orphans commands in usage
Fixes: #14516
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked from commit 105a76bf542e05b739d5a03ca8ae55432350f107)
Conflicts:
src/rgw/rgw_admin.cc (trivial resolution)
commit 0e1378effdd1d0d70d3de05c79b208e9f8b8e328
Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
Date: Mon Sep 21 20:32:29 2015 +0200
[backport] rgw: fix wrong etag calculation during POST on S3 bucket.
Closes: #14570
(cherry picked from commit 742906a)
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
Signed-off-by: Vladislav Odintsov <odivlad@gmail.com>
Tested-by: Vladislav Odintsov <odivlad@gmail.com>
commit 5c8d1d74069f70b85bc4286e6d1136bce1dc593f
Author: Yuan Zhou <yuan.zhou@intel.com>
Date: Thu Jul 9 16:56:07 2015 +0800
[backport] rgw: Make RGW_MAX_PUT_SIZE configurable
The 5GB limit of a single operation uploading was part of S3 spec.
However some private setups may have some special requirements
on this limit. It's more convinent to have a configurable value.
Closes: http://tracker.ceph.com/issues/14569
(cherry picked from commit df97f28)
Signed-off-by: Yuan Zhou <yuan.zhou@intel.com>
Signed-off-by: Vladislav Odintsov <odivlad@gmail.com>
Tested-by: Vladislav Odintsov <odivlad@gmail.com>
commit f2ca42b1218182f4bfa27718c9606705d8b9941f
Author: Ken Dreyer <kdreyer@redhat.com>
Date: Wed Feb 3 19:51:58 2016 -0700
doc: add orphans commands to radosgw-admin(8)
The --help text was added in 105a76bf542e05b739d5a03ca8ae55432350f107.
Add the corresponding entries to the man page.
Fixes: #14637
Signed-off-by: Ken Dreyer <kdreyer@redhat.com>
(cherry picked from commit ec162f068b40f594c321df5caa9fe2541551b89e)
Cherry-pick to hammer includes nroff source change (in master the
nroff sources are no longer present in Git.)
commit e42ed6d6414ad55d671dd0f406b1dababd643af8
Author: Ken Dreyer <kdreyer@redhat.com>
Date: Thu Feb 4 11:04:39 2016 -0700
man: rebuild manpages
following the procedure in admin/manpage-howto.txt.
Signed-off-by: Ken Dreyer <kdreyer@redhat.com>
commit a8fc6a9ffb2f8021657ad412fd0aaaaf7f98bd53
Author: Greg Farnum <gfarnum@redhat.com>
Date: Wed Jan 13 13:17:53 2016 -0800
fsx: checkout old version until it compiles properly on miras
I sent a patch to xfstests upstream at
http://article.gmane.org/gmane.comp.file-systems.fstests/1665, but
until that's fixed we need a version that works in our test lab.
Signed-off-by: Greg Farnum <gfarnum@redhat.com>
(cherry picked from commit 7d52372ae74878ebd001036ff0a7aad525eb15b6)
commit f3bab8c7bc6bba3c79329d7106684596634e17a6
Merge: 31d86b1 1b02859
Author: Loic Dachary <loic@dachary.org>
Date: Wed Feb 3 12:41:56 2016 +0700
Merge pull request #7454 from dachary/wip-14584-hammer
hammer: fsstress.sh fails
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit 31d86b1580d59581d130e7a5e84905a5b4d67666
Merge: 2c8e579 2817ffc
Author: Loic Dachary <loic@dachary.org>
Date: Wed Feb 3 11:38:57 2016 +0700
Merge pull request #6918 from asheplyakov/hammer-bug-12449
osd: check for full before changing the cached obc (hammer)
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit 1b02859480677abbd8708650764ed6815917e0cd
Author: Yan, Zheng <zyan@redhat.com>
Date: Tue Aug 18 15:22:55 2015 +0800
qa/fsstress.sh: fix 'cp not writing through dangling symlink'
On some test machines, /usr/lib/ltp/testcases/bin/fsstress is
dangling symlink. 'cp -f' is impotent in this case.
Fixes: #12710
Signed-off-by: Yan, Zheng <zyan@redhat.com>
(cherry picked from commit 479f2a760baf6af4714d3391a366a6b3acf1bcaf)
commit 2c8e57934284dae0ae92d1aa0839a87092ec7c51
Merge: 1cab151 700be56
Author: Sage Weil <sage@redhat.com>
Date: Sat Jan 30 21:42:29 2016 -0500
Merge pull request #7236 from athanatos/wip-14376
config_opts: increase suicide timeout to 300 to match recovery
http://pulpito.ceph.com/sage-2016-01-30_09:58:32-rados-wip-sage-testing-hammer---basic-mira/
Reviewed-by: Sage Weil <sage@redhat.com>
commit 1cab151a6cdc8a1b9b2f11518b77ea149c19d067
Merge: a134c44 5105d50
Author: Sage Weil <sage@redhat.com>
Date: Sat Jan 30 21:42:12 2016 -0500
Merge pull request #6450 from dachary/wip-13672-hammer
tests: testprofile must be removed before it is re-created
http://pulpito.ceph.com/sage-2016-01-30_09:58:32-rados-wip-sage-testing-hammer---basic-mira/
Reviewed-by: Sage Weil <sage@redhat.com>
commit a134c44d626ead3308464474d67604363bac6f5d
Merge: 520792d 3f1292e
Author: Sage Weil <sage@redhat.com>
Date: Sat Jan 30 21:41:39 2016 -0500
Merge pull request #6680 from SUSE/wip-13859-hammer
hammer: ceph.spec.in License line does not reflect COPYING
http://pulpito.ceph.com/sage-2016-01-30_09:58:32-rados-wip-sage-testing-hammer---basic-mira/
commit 520792d51e949c451767d116e267d86fee812ada
Merge: 4d0fafb c2c6d02
Author: Sage Weil <sage@redhat.com>
Date: Sat Jan 30 21:41:18 2016 -0500
Merge pull request #6791 from branch-predictor/bp-5812-backport
Race condition in rados bench
http://pulpito.ceph.com/sage-2016-01-30_09:58:32-rados-wip-sage-testing-hammer---basic-mira/
commit 4d0fafb289fc35f44e6e74bb974c402ba147d4d4
Merge: 211a093 6379ff1
Author: Sage Weil <sage@redhat.com>
Date: Sat Jan 30 21:40:38 2016 -0500
Merge pull request #6973 from dreamhost/wip-configure-hammer
configure.ac: no use to add "+" before ac_ext=c
http://pulpito.ceph.com/sage-2016-01-30_09:58:32-rados-wip-sage-testing-hammer---basic-mira/
Reviewed-by: Kefu Chai <kchai@redhat.com>
commit 211a093d4107c1806ad7e4876bc5c550a5b5b7d4
Merge: 0c13656 53742bd
Author: Sage Weil <sage@redhat.com>
Date: Sat Jan 30 21:40:13 2016 -0500
Merge pull request #7206 from dzafman/wip-14292
osd/PG.cc: 3837: FAILED assert(0 == "Running incompatible OSD")
http://pulpito.ceph.com/sage-2016-01-30_09:58:32-rados-wip-sage-testing-hammer---basic-mira/
Reviewed-by: David Zafman <dzafman@redhat.com>
commit 0c136561600e295ec48dcf29a77aa2cd293a7236
Merge: 1ea14ba ae56de0
Author: Sage Weil <sage@redhat.com>
Date: Sat Jan 30 21:39:42 2016 -0500
Merge pull request #7207 from rldleblanc/recency_fix_for_hammer
hammer: osd/ReplicatedPG: Recency fix for Hammer
Reviewed-by: Sage Weil <sage@redhat.com>
commit 1ea14bae10930ed3e66b7b0140e359009a44275e
Merge: 1740d8c 8d9e08c
Author: Sage Weil <sage@redhat.com>
Date: Sat Jan 30 21:39:11 2016 -0500
Merge pull request #7347 from tchaikov/wip-hammer-10093
tools: ceph-monstore-tool must do out_store.close()
http://pulpito.ceph.com/sage-2016-01-30_09:58:32-rados-wip-sage-testing-hammer---basic-mira/
Reviewed-by: Sage Weil <sage@redhat.com>
commit 1740d8ce8eb7e7debce1684a19736783489628c3
Merge: 7848cdc c7252a3
Author: Sage Weil <sage@redhat.com>
Date: Sat Jan 30 21:38:35 2016 -0500
Merge pull request #7411 from dachary/wip-14467-hammer
hammer: disable filestore_xfs_extsize by default
http://pulpito.ceph.com/sage-2016-01-30_09:58:32-rados-wip-sage-testing-hammer---basic-mira/
Reviewed-by: Sage Weil <sage@redhat.com>
commit 7848cdc4c8c3598cd291d26a2dd4d76abc4bcda9
Merge: 57abeab 70f1ba3
Author: Sage Weil <sage@redhat.com>
Date: Sat Jan 30 21:38:13 2016 -0500
Merge pull request #7412 from dachary/wip-14470-hammer
tools: tool for artificially inflate the leveldb of the mon store for testing purposes
Reviewed-by: Sage Weil <sage@redhat.com>
commit 57abeab1f2610d89f0ab2a028c8e093afde5e403
Merge: 4d7d7c3 9109304
Author: Sage Weil <sage@redhat.com>
Date: Sat Jan 30 21:37:46 2016 -0500
Merge pull request #7446 from liewegas/wip-14537-hammer
mon: compact full epochs also
http://pulpito.ceph.com/sage-2016-01-30_09:58:32-rados-wip-sage-testing-hammer---basic-mira/
Reviewed-by: Sage Weil <sage@redhat.com>
commit 4d7d7c33ff58ffeb27b47d379bf7dd4af90785b0
Merge: 8360486 8c28f2f
Author: Josh Durgin <jdurgin@redhat.com>
Date: Sat Jan 30 11:45:31 2016 -0800
Merge pull request #7182 from dachary/wip-14143-hammer
hammer: Verify self-managed snapshot functionality on image create
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
commit 8360486764c3ab4e10a34f5bdf7555e4c3385977
Merge: 501e01a b2961ce
Author: Josh Durgin <jdurgin@redhat.com>
Date: Sat Jan 30 11:45:20 2016 -0800
Merge pull request #7183 from dachary/wip-14283-hammer
hammer: rbd: fix bench-write
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
commit 501e01acc55f846cd3ad85e8294a9598c1c90907
Merge: 97d4f6a 24c0b27
Author: Josh Durgin <jdurgin@redhat.com>
Date: Sat Jan 30 11:45:05 2016 -0800
Merge pull request #7416 from dachary/wip-14466-hammer
hammer: rbd-replay does not check for EOF and goes to endless loop
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
commit 97d4f6a7bed5a0d47f18b3a223a916aef355eaf9
Merge: 51cc015 46d626d
Author: Josh Durgin <jdurgin@redhat.com>
Date: Sat Jan 30 11:44:50 2016 -0800
Merge pull request #7417 from dachary/wip-14553-hammer
hammer: rbd: TaskFinisher::cancel should remove event from SafeTimer
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
commit 51cc015c7e20d2ea9b9517238481fb80687be17c
Merge: 23c24fc 73e03de
Author: Josh Durgin <jdurgin@redhat.com>
Date: Sat Jan 30 11:44:32 2016 -0800
Merge pull request #7407 from dillaman/wip-14543-hammer
librbd: ImageWatcher shouldn't block the notification thread
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
commit 23c24fc50d07a07db12239ac372bfce5a9fe2255
Merge: 9513391 26e832e
Author: Josh Durgin <jdurgin@redhat.com>
Date: Sat Jan 30 11:44:12 2016 -0800
Merge pull request #6980 from dillaman/wip-14063-hammer
librbd: fix merge-diff for >2GB diff-files
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
commit 91093041a86fda40de08a366b5118e5e3ae275f0 (refs/remotes/me/wip-14537-hammer)
Author: Kefu Chai <kchai@redhat.com>
Date: Thu Jan 28 02:09:53 2016 -0800
mon: compact full epochs also
by compacting the ${prefix}.${start}..${prefix}..${end} does not
necessary compact the range of ${prefix}."full_"${start}..
${prefix}."full_"${end}. so when more and more epochs get trimmed
with out a full range compaction, the size of monitor store could
be very large.
Fixes: #14537
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 93d633a25ad8c36c972bb766c38187e2612041e1)
commit 2817ffcf4e57f92551b86388681fc0fe70c386ec
Author: Alexey Sheplyakov <asheplyakov@mirantis.com>
Date: Wed Dec 16 15:31:52 2015 +0300
Check for full before changing the cached obc
ReplicatedPG::prepare_transaction(): check if the pool is full before
updating the cached ObjectContext to avoid the discrepancy between
the cached and the actual object size (and other metadata).
While at it improve the check itself: consider cluster full flag,
not just the pool full flag, also consider object count changes too,
not just bytes.
Based on commit a1eb380c3d5254f9f1fe34b4629e51d77fe010c1
Fixes: #13335
Signed-off-by: Alexey Sheplyakov <asheplyakov@mirantis.com>
commit 951339103d35bc8ee2de880f77aada40d15b592a (refs/remotes/gh/wip-test-14716-4)
Merge: e43aca5 5e5b512
Author: Loic Dachary <loic@dachary.org>
Date: Fri Jan 29 23:31:47 2016 +0700
Merge pull request #6353 from theanalyst/wip-13513-hammer
rgw: value of Swift API's X-Object-Manifest header is not url_decoded during segment look up
Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com>
commit e43aca51d630274a57237b0f91a56df79ce8874a
Merge: 7e20e6e 4420929
Author: Loic Dachary <loic@dachary.org>
Date: Fri Jan 29 23:31:16 2016 +0700
Merge pull request #6620 from SUSE/wip-13820-hammer
hammer: rgw: Setting ACL on Object removes ETag
Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com>
commit 7e20e6e9d6626c5ac8b1f479011ab77a9e87da23
Merge: f1d5570 cbb5c1f
Author: Loic Dachary <loic@dachary.org>
Date: Fri Jan 29 23:30:57 2016 +0700
Merge pull request #7186 from dachary/wip-13888-hammer
hammer: rgw: orphans finish segfaults
Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com>
commit ae56de09fe1385958b5b600d1f0c91383989926f
Author: Sage Weil <sage@redhat.com>
Date: Wed Nov 25 14:40:26 2015 -0500
osd: recency should look at newest (not oldest) hitsets
Reported-by: xinxin shu <xinxin.shu@intel.com>
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 528eae92d010ea34ae8cf0e8b2290aaa5e058d24)
Conflicts:
src/osd/ReplicatedPG.cc
code section was moved to ReplicatedPG::maybe_promote
in master.
Signed-off-by: Robert LeBlanc <robert.leblanc@endurance.com>
commit 5cefcb975771f0c2efb7dfc77ce14a93a4ee7f1b
Author: Sage Weil <sage@redhat.com>
Date: Wed Nov 25 14:39:08 2015 -0500
osd/ReplicatedPG: fix promotion recency logic
Recency is defined as how many of the last N hitsets an object
must appear in in order to be promoted. The previous logic did
nothing of the sort... it checked for the object in any one of
the last N hitsets, which led to way to many promotions and killed
any chance of the cache performing properly.
While we are here, we can simplify the code to drop the max_in_*
fields (no longer necessary).
Note that we may still want a notion of 'temperature' that does
tolerate the object missing in one of the recent hitsets.. but
that would be different than recency, and should probably be
modeled after the eviction temperature model.
Backport: infernalis, hammer
Reported-by: Nick Fisk <nick@fisk.me.uk>
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 180c8743addc5ae2f1db9c58cd2996ca6e7ac18b)
Conflicts:
src/osd/ReplicatedPG.cc
code section was moved to ReplicatedPG::maybe_promote
in master.
Signed-off-by: Robert LeBlanc <robert.leblanc@endurance.com>
commit f1d5570beab0769b925b917e402d441ff053794c
Merge: c4bb343 50c82f2
Author: Sage Weil <sage@redhat.com>
Date: Fri Jan 29 08:52:51 2016 -0500
Merge pull request #5789 from SUSE/wip-12928-hammer
rpm: libcephfs_jni1 has no %post and %postun
Reviewed-by: Sage Weil <sage@redhat.com>
commit c4bb34320df8bfa734512a400fe8664c131363ff
Merge: 86ba6ca a5e4f70
Author: Sage Weil <sage@redhat.com>
Date: Fri Jan 29 08:50:56 2016 -0500
Merge pull request #7434 from tchaikov/wip-14441-hammer
man: document listwatchers cmd in "rados" manpage
Reviewed-by: Sage Weil <sage@redhat.com>
commit a5e4f70af422b05ece53f245bc15491bb1dd540d
Author: Kefu Chai <kchai@redhat.com>
Date: Wed Dec 23 11:23:38 2015 +0800
man: document listwatchers cmd in "rados" manpage
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit c2e391301efc43f0b431e89737246b2c43bf10a9)
man/rados.8: also added the rendered man.8 man page, as we don't
put the generated man pages in master anymore. but
they are still in the hammer's source repo.
commit 46d626d92d7af218816d831cfcc1265b3ea31b72
Author: Douglas Fuller <dfuller@redhat.com>
Date: Fri Jan 22 11:18:40 2016 -0800
rbd: remove canceled tasks from timer thread
When canceling scheduled tasks using the timer thread, TaskFinisher::cancel
does not call SafeTimer::cancel_event, so events fire anyway. Add this call.
Fixes: #14476
Signed-off-by: Douglas Fuller <dfuller@redhat.com>
(cherry picked from commit 2aa0f318c862dbe3027d74d345671506605778eb)
commit 24c0b27c6f6a26c2b7bab5bcbc421a18592d026f
Author: Mykola Golub <mgolub@mirantis.com>
Date: Thu Jan 21 13:45:42 2016 +0200
rbd-replay: handle EOF gracefully
Fixes: #14452
Signed-off-by: Mykola Golub <mgolub@mirantis.com>
(cherry picked from commit c59b84c3e2c9bbda68219e4d2288a889dd9ca6cb)
Conflicts:
src/rbd_replay/BufferReader.cc: trivial resolution
caused by s/CEPH_PAGE_SIZE/CEPH_BUFFER_APPEND_SIZE/
in the context of one hunk.
commit 70f1ba33fe1a81d2631d54429749433a6cbfca44
Author: Cilang Zhao <zhao.cilang@h3c.com>
Date: Tue Jan 5 14:34:05 2016 +0800
tools: monstore: add 'show-versions' command.
Using this tool, the first/last committed version of maps will be shown.
Signed-off-by: Cilang Zhao <zhao.cilang@h3c.com>
(cherry picked from commit 21e6ba0c18428caff45733e6b43d197be38af8bb)
commit 926017187910c9e6a3fb8babf9b498cf07941819
Author: Kefu Chai <kchai@redhat.com>
Date: Wed Sep 16 18:28:52 2015 +0800
tools: ceph_monstore_tool: add inflate-pgmap command
this command repeatly add the latest pgmap to the monstore in order
to inflate it. the command helps with the testing of some monstore
related performance issue of monitor
Fixes: #14217
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit b0f6af814a58dd374ca95e84a4e381f8ef1368de)
commit a1d5728c928eb7e6b8c655741a0db3398ced7d56
Author: Bo Cai <cai.bo@h3c.com>
Date: Tue Oct 20 15:23:49 2015 +0800
tools:support printing the crushmap in readable fashion.
Signed-off-by: Bo Cai <cai.bo@h3c.com>
(cherry picked from commit b7faf678ed0241abc7eb86b9baaa6db669a22d53)
commit bd9484750f546974047dacd9176a8072be846596
Author: Bo Cai <cai.bo@h3c.com>
Date: Mon Sep 14 19:50:47 2015 +0800
tools:print the map infomation in human readable format.
Signed-off-by: Bo Cai <cai.bo@h3c.com>
(cherry picked from commit 5ed8cdc19150382c946a373ec940d76f98e6ecb7)
commit fba65832aad8a46d94a9256a56997e9df9e62297
Author: Bo Cai <cai.bo@h3c.com>
Date: Mon Sep 14 19:19:05 2015 +0800
tools:remove the local file when get map failed.
Signed-off-by: Bo Cai <cai.bo@h3c.com>
(cherry picked from commit 0b03b32d8ba76fe9f6f1158e68eb440e3670393a)
commit 1bb899a290b77188b44a53ef7c7a40910c9248b2
Author: Joao Eduardo Luis <joao@suse.de>
Date: Mon Jul 13 12:35:13 2015 +0100
tools: ceph_monstore_tool: describe behavior of rewrite command
Signed-off-by: Joao Eduardo Luis <joao@suse.de>
(cherry picked from commit a881f9385feb0f5a61fa22357984d6f291c08177)
commit 9035c69481f4aa4786414ac41cbc36fb4a3ca51d
Author: Kefu Chai <kchai@redhat.com>
Date: Fri Jun 19 22:57:57 2015 +0800
tools/ceph-monstore-tools: add rewrite command
"rewrite" command will
- add a new osdmap version to update current osdmap held by OSDMonitor
- add a new paxos version, as a proposal it will
* rewrite all osdmap epochs from specified epoch to the last_committed
one with the specified crush map.
* add the new osdmap which is added just now
so the leader monitor can trigger a recovery process to apply the transaction
to all monitors in quorum, and hence bring them back to normal after being
injected with a faulty crushmap.
Fixes: #11815
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 306373427836ca0c2418dbe6caab26d74d94d12e)
commit 90aaed744243dfc7c620f91e19fc0bfa233f711f
Author: huangjun <hjwsm1989@gmail.com>
Date: Tue Apr 21 14:04:40 2015 +0800
tools: ceph-monstore-tool must do out_store.close()
this change fixes the "store-copy" command.
Like the bug reported in http://tracker.ceph.com/issues/10093.
Signed-off-by: huangjun <hjwsm1989@gmail.com>
(cherry picked from commit d85e0f8c50fce62be012506f7ffcb1cdeb0c819c)
commit c7252a3d54bae88f91c2b4e63fc9c27cfbb2423e
Author: Ken Dreyer <kdreyer@redhat.com>
Date: Mon Jan 18 08:24:46 2016 -0700
osd: disable filestore_xfs_extsize by default
This option involves a tradeoff: When disabled, fragmentation is worse,
but large sequential writes are faster. When enabled, large sequential
writes are slower, but fragmentation is reduced.
Fixes: #14397
(cherry picked from commit aed85775bf53c273786ce4999320134822722af5)
commit 86ba6caf992d2544cdd174e3b3f26a6099c91fc5
Merge: b6b8ee4 0325f8a
Author: Loic Dachary <loic@dachary.org>
Date: Fri Jan 29 10:36:05 2016 +0700
Merge pull request #7316 from ceph/wip-deb-lttng-hammer
deb: strip tracepoint libraries from Wheezy/Precise builds
All other "modern" Debian-based OSes have a functional LTTng-UST. Since only hammer needs to build on these older distros, this fix only affects the deb building process for those two releases(since autoconf detects that LTTng is broken).
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit b6b8ee41fc77679e82304e79da6dbd0c35f5c067
Merge: d54840b f96c812
Author: Loic Dachary <loic@dachary.org>
Date: Fri Jan 29 10:21:43 2016 +0700
Merge pull request #7187 from dachary/wip-13831-hammer
hammer: init script reload doesn't work on EL7
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit 73e03def9271fb5d1739b195e428c3ebfcebd59b
Author: Jason Dillaman <dillaman@redhat.com>
Date: Thu Jan 28 14:38:20 2016 -0500
librbd: ImageWatcher shouldn't block the notification thread
Blocking the notification thread will also result in librados async
callbacks becoming blocked (since they use the same thread).
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 7e2019a72733dff43e55c9b22df12939d584f87d)
Conflicts:
src/librbd/ImageWatcher.[cc|h]: fewer RPC messages
commit 5617166f78c1995436b4e0794dab2d8254331815
Author: Jason Dillaman <dillaman@redhat.com>
Date: Thu Jan 28 14:35:54 2016 -0500
librados_test_stub: watch/notify now behaves similar to librados
Notifications are executed via the same librados AIO callback
thread, so it's now possible to catch deadlock.
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 43722571838a2d78ca8583b5a1ea381cd988de0b)
commit 8fc82b23304ef327933723373cd4d1090d04bfbc
Author: Jason Dillaman <dillaman@redhat.com>
Date: Thu Jan 28 12:40:18 2016 -0500
tests: simulate writeback flush during snap create
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit e9570fcf4be9cd5c06937769d074fa52ffb32e4b)
commit d54840bf4a70fc65285bbfdff0c7bf8f579643b1
Merge: 4051bc2 9f30fe1
Author: Sage Weil <sage@redhat.com>
Date: Tue Jan 26 10:12:29 2016 -0500
Merge pull request #7365 from liewegas/wip-tcmalloc-hammer
osd: pass tcmalloc env through to ceph-osd
Reviewed-by: Ken Dreyer <kdreyer@redhat.com>
commit 9f30fe18e839f5846c90e3b5995171a0132d7f3a
Author: Sage Weil <sage@redhat.com>
Date: Tue Jan 26 08:43:15 2016 -0500
upstart/ceph-osd.conf: pass TCMALLOC_MAX_TOTAL_THREAD_CACHE_BYTES through
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit a95a3d34cccb9f9f0782a96ca05e5fe9c2b01772)
commit a58873e213d7423d89c95db4e1710dc9631e3313
Author: Sage Weil <sage@redhat.com>
Date: Wed Jan 20 18:36:08 2016 -0500
init-ceph: pass TCMALLOC_MAX_TOTAL_THREAD_CACHE_BYTES through
..when set in the sysconfig/default file.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 03e01ae263ef207046933890821ae75d5e60d9b8)
[source default/sysconfig file too]
commit 8d9e08c44150a4f3ad06ec1abc130b9aeaf479d9
Author: huangjun <hjwsm1989@gmail.com>
Date: Tue Apr 21 14:04:40 2015 +0800
tools: ceph-monstore-tool must do out_store.close()
this change fixes the "store-copy" command.
Like the bug reported in http://tracker.ceph.com/issues/10093.
Signed-off-by: huangjun <hjwsm1989@gmail.com>
(cherry picked from commit d85e0f8c50fce62be012506f7ffcb1cdeb0c819c)
commit 0325f8af5cbee3d74e9f363f61c2e2ababf501d9
Author: Jason Dillaman <dillaman@redhat.com>
Date: Thu Jan 21 15:36:37 2016 -0500
deb: strip tracepoint libraries from Wheezy/Precise builds
These releases do not (by default) have a sane LTTng-UST environment,
which results in autoconf disabling support for tracing.
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
commit 4051bc2a5e4313ac0f6236d7a34ed5fb4a1d9ea2
Merge: 4e67418 e4d3e9b
Author: Jason Dillaman <dillaman@redhat.com>
Date: Mon Jan 18 20:46:07 2016 -0500
Merge pull request #7252 from ceph/wip-13483-hammer
qa: remove legacy OS support from rbd/qemu-iotests
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
commit e4d3e9b29aa6a5f9efff8b787949a5bbf48ad7de (refs/remotes/gh/wip-13483-hammer)
Author: Jason Dillaman <dillaman@redhat.com>
Date: Mon Jan 18 15:57:43 2016 -0500
qa: disable rbd/qemu-iotests test case 055 on RHEL/CentOS
Fixes: #14385
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
commit f7acd44c26310242b69ee50322bd6b43fdc774b9
Author: Jason Dillaman <dillaman@redhat.com>
Date: Thu Oct 15 12:44:06 2015 -0400
qa: remove legacy OS support from rbd/qemu-iotests
RHEL7 derivatives were failing test 002 since they were using
legacy test cases for now unsupported OSes.
Fixes: #13483
Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
commit 8c28f2f28d960d823ffd632671edaf029c30fb0f
Author: Jason Dillaman <dillaman@redhat.com>
Date: Mon Dec 14 17:41:49 2015 -0500
librbd: optionally validate RBD pool configuration (snapshot support)
Fixes: #13633
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 1fea4dadc60e13518e9ee55d136fbc4e9d3a621e)
Conflicts:
src/common/config_opts.h: trivial resolution
commit 700be56c530879a72a628c62265d18f0a5d8fb3b
Author: Samuel Just <sjust@redhat.com>
Date: Thu Jan 14 08:35:23 2016 -0800
config_opts: increase suicide timeout to 300 to match recovery
Fixes: 14376
Backport: hammer, firefly
Signed-off-by: Samuel Just <sjust@redhat.com>
commit 4e67418958e5caf5e4f81c4ed566e8c7269930fa
Merge: 28e99a9 1ab2b48
Author: Sage Weil <sage@redhat.com>
Date: Thu Jan 14 09:21:50 2016 -0500
Merge pull request #7179 from dachary/wip-14287-hammer
hammer: ReplicatedPG: wrong result code checking logic during sparse_read
Reviewed-by: Sage Weil <sage@redhat.com>
commit 28e99a90b38ce8c0285c1aec9c5524365385be00
Merge: 17a2965 86f5cf6
Author: Sage Weil <sage@redhat.com>
Date: Thu Jan 14 09:20:53 2016 -0500
Merge pull request #7178 from dachary/wip-14285-hammer
hammer: osd/OSD.cc: 2469: FAILED assert(pg_stat_queue.empty()) on shutdown
Reviewed-by: Sage Weil <sage@redhat.com>
commit 17a2965825494765249c244aef59ebd821711d42
Merge: a1459ea b0856ee
Author: Sage Weil <sage@redhat.com>
Date: Thu Jan 14 09:20:34 2016 -0500
Merge pull request #7177 from dachary/wip-14043-hammer
hammer: osd/PG.cc: 288: FAILED assert(info.last_epoch_started >= info.history.last_epoch_started)
Reviewed-by: Sage Weil <sage@redhat.com>
commit a1459eacd7cc0d064493d01f963b8c1231cd3518
Merge: c9a13a1 9cee89b
Author: Sage Weil <sage@redhat.com>
Date: Thu Jan 14 09:20:10 2016 -0500
Merge pull request #7180 from dachary/wip-14288-hammer
hammer: ceph osd pool stats broken in hammer
Reviewed-by: Sage Weil <sage@redhat.com>
commit c9a13a12e88c9a18bfb3cc99d4a7c103b4330426
Merge: 174a1a9 4d0b9a1
Author: Sage Weil <sage@redhat.com>
Date: Thu Jan 14 09:19:18 2016 -0500
Merge pull request #6994 from badone/wip-13993-hammer
log: Log.cc: Assign LOG_DEBUG priority to syslog calls
Reviewed-by: Sage Weil <sage@redhat.com>
commit 174a1a974725ce4025be4dfdf3b13df766dfac76
Merge: cb167e9 6cf261c
Author: Sage Weil <sage@redhat.com>
Date: Thu Jan 14 09:18:55 2016 -0500
Merge pull request #6839 from SUSE/wip-13789-hammer
Objecter: potential null pointer access when do pool_snap_list.
Reviewed-by: Sage Weil <sage@redhat.com>
commit cb167e998bbbdf1b9c273f6bc4f118796d396d37
Merge: 1c4ba85 66ff0aa
Author: Sage Weil <sage@redhat.com>
Date: Thu Jan 14 09:18:23 2016 -0500
Merge pull request #6835 from SUSE/wip-13892-hammer
hammer: auth/cephx: large amounts of log are produced by osd
Reviewed-by: Sage Weil <sage@redhat.com>
commit 1c4ba855b34290069c7ce0fa990fe72b7a1b381e
Merge: 72b6b68 61da13b
Author: Sage Weil <sage@redhat.com>
Date: Thu Jan 14 09:17:53 2016 -0500
Merge pull request #6834 from SUSE/wip-13930-hammer
hammer: Ceph Pools' MAX AVAIL is 0 if some OSDs' weight is 0
Reviewed-by: Sage Weil <sage@redhat.com>
commit 72b6b68f43ffc953ca7f3056574bd68428a5abe8
Merge: d064f9f 53e81aa
Author: Sage Weil <sage@redhat.com>
Date: Thu Jan 14 09:17:25 2016 -0500
Merge pull request #6832 from SUSE/wip-13936-hammer
hammer: Ceph daemon failed to start, because the service name was already used.
Reviewed-by: Sage Weil <sage@redhat.com>
commit d064f9f6b3bc93fa5887304a9b9ab92d503ebd15
Merge: 0ca6285 ac05617
Author: Sage Weil <sage@redhat.com>
Date: Thu Jan 14 09:16:54 2016 -0500
Merge pull request #6755 from SUSE/wip-13870-hammer
hammer: OSD: race condition detected during send_failures
Reviewed-by: Sage Weil <sage@redhat.com>
commit 0ca62855f22136c7042a5988366d46e59629f206
Merge: e9f545b a81bcf7
Author: Loic Dachary <loic@dachary.org>
Date: Thu Jan 14 08:22:35 2016 +0100
Merge pull request #6415 from dillaman/wip-13541-hammer
LTTng-UST tracing should be dynamically enabled
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit e9f545b4a44c91faf91a8ae3824b1b0d46909244
Merge: 8521916 b2f1e76
Author: Loic Dachary <loic@dachary.org>
Date: Thu Jan 14 08:15:02 2016 +0100
Merge pull request #7176 from dachary/wip-13440-hammer
ceph-disk prepare fails if device is a symlink
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit 85219163f9b78faa3c1fe96cad7129c6813550c0
Merge: ebfb3fb 5264bc6
Author: Loic Dachary <loic@dachary.org>
Date: Thu Jan 14 08:14:31 2016 +0100
Merge pull request #7150 from jecluis/wip-14236
mon: OSDMonitor: do not assume a session exists in send_incremental()
Reviewed-by: Kefu Chai <kchai@redhat.com>
commit ebfb3fbe8a1920474941d4a95540cc4a53a747d4
Merge: e1b9208 447ab1d
Author: Josh Durgin <jdurgin@redhat.com>
Date: Wed Jan 13 17:15:32 2016 -0800
Merge pull request #7226 from dillaman/wip-13810-hammer
tests: notification slave needs to wait for master
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
commit 447ab1d70a63fceed39518d5714c138c54f6525e
Author: Jason Dillaman <dillaman@redhat.com>
Date: Wed Jan 13 12:44:01 2016 -0500
tests: notification slave needs to wait for master
If the slave instance starts before the master, race
conditions are possible.
Fixes: #13810
Backport: infernalis, hammer
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 3992d6fe67bbf82322cedc1582406caaf6d4de60)
commit 53742bdbc587747c93413f9db38f3eeb71487872
Author: David Zafman <dzafman@redhat.com>
Date: Mon Jun 15 17:55:41 2015 -0700
ceph_osd: Add required feature bits related to this branch to osd_required mask
Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 0b2bab460c013ea5cece10ea028d26da3485eaf5)
commit 3066231865e2fe56344de9db26024ac65e03053d
Author: David Zafman <dzafman@redhat.com>
Date: Thu Jun 4 18:47:42 2015 -0700
osd: CEPH_FEATURE_CHUNKY_SCRUB feature now required
Feature present since at least the Dumpling release.
A later commit will add it to the osd_required mask
Fixes: #11661
Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 7a10a7e196efd0f59fa7adb87d0a120655b227d8)
commit e1b92081c9e4b21eb30cc873c239083a08fce12f
Merge: 9708e59 3d3595f
Author: Loic Dachary <loic@dachary.org>
Date: Tue Jan 12 14:28:26 2016 +0100
Merge pull request #6530 from SUSE/wip-13760-hammer
unknown argument --quiet in udevadm settle
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit 6379ff16dbc7fdecccaa8874d978d1ab58ce44cf
Author: Kefu Chai <kchai@redhat.com>
Date: Tue May 5 15:07:33 2015 +0800
configure.ac: no use to add "+" before ac_ext=c
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 899dd23766c5ae6fef39bf24ef0692127d17deb3)
Signed-off-by: Robin H. Johnson <robin.johnson@dreamhost.com>
commit f96c8125d423f24e52df0f8420b0b2748e34ca90
Author: Herve Rousseau <hroussea@cern.ch>
Date: Fri Nov 6 09:52:28 2015 +0100
rgw: fix reload on non Debian systems.
When using reload in non-debian systems, /bin/sh's kill is used to send the HUP signal to the radosgw process.
This kill version doesn't understand -SIGHUP as a valid signal, using -HUP does work.
Fix: #13709
Backport: hammer
Signed-off-by: Hervé Rousseau <hroussea@cern.ch>
(cherry picked from commit 1b000abac3a02d1e788bf25eead4b6873133f5d2)
commit cbb5c1fc2386205277ad22474c1f696a07fcb972
Author: Igor Fedotov <ifedotov@mirantis.com>
Date: Thu Nov 19 13:38:40 2015 +0300
Fixing NULL pointer dereference
Signed-off-by: Igor Fedotov <ifedotov@mirantis.com>
(cherry picked from commit 93d3dfe0441be50a6990d458ee0ee3289af39b20)
commit b2961cef3d8e84d2302815e33eb7dc9033d2bb78
Author: Sage Weil <sage@redhat.com>
Date: Tue Aug 18 16:05:29 2015 -0400
rbd: fix bench-write
Make each IO get a unique offset!
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 333f3a01a9916c781f266078391c580efb81a0fc)
commit 9cee89bbdd1d89d2fe72c8d73fc0c18db515198a
Author: BJ Lougee <almightybeeij@gmail.com>
Date: Mon Apr 13 11:24:38 2015 -0500
Check that delta_sum.stats.sum.num_object_copies and delta_sum.stats.sum.num_object are greater than zero
This fixes division by zero.
Signed-off-by BJ Lougee <almightybeeij@gmail.com>
(cherry picked from commit 27ed729c1088133400aa072eeca9e125942f2d94)
commit 1ab2b481596ad5296e168bf8027e59d41190176e
Author: xiexingguo <258156334@qq.com>
Date: Tue Dec 22 17:05:06 2015 +0800
ReplicatedPG: fix sparse-read result code checking logic
Move ahead the result code checking logic before we continue to verify the trailing hole, otherwise
the real result of non-hole reading may be overwritten and thus confuse caller.
Fixes: #14151
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
(cherry picked from commit b3aa29e324cf0a96a1f2f5dcf1ba998219457bcd)
Conflicts:
src/osd/ReplicatedPG.cc: trivial resolution
commit 86f5cf6caa36760802775df19cdabe55bcafa33e
Author: Sage Weil <sage@redhat.com>
Date: Fri Jan 1 09:32:04 2016 -0500
osd: clear pg_stat_queue after stopping pgs
Fixes: #14212
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit c25ff9927febe77cda31b534971b62f75c196ca2)
commit b0856ee67c0e7cf6ab6095d3f657c18014859526
Author: David Zafman <dzafman@redhat.com>
Date: Thu Dec 3 14:52:24 2015 -0800
osd: Test osd_find_best_info_ignore_history_les config in another assert
Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 02a9a41f151a3d968bf8066749658659dc6e3ac4)
commit b2f1e76d8e95b6e2f88fa1c122ea8cd24532cd10
Author: Joe Julian <jjulian@io.com>
Date: Fri Oct 9 12:57:06 2015 -0700
Compare parted output with the dereferenced path
Compare parted output with the dereferenced path of the device as parted
prints that instead of the symlink we called it with.
http://tracker.ceph.com/issues/13438 Fixes: #13438
Signed-off-by: Joe Julian <jjulian@io.com>
(cherry picked from commit b3c7cb098195111b9c642e5a9b726b63717f2e0d)
commit 9708e59a1fc2ae52cfae848ce585751bc9fbe572
Merge: 9739d4d b62cac6
Author: Loic Dachary <loic@dachary.org>
Date: Mon Jan 11 09:01:12 2016 +0100
Merge pull request #6512 from SUSE/wip-13734-hammer
rgw: swift API returns more than real object count and bytes used when retrieving account metadata
Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com>
commit 5264bc677f0f612165987bf51fe7d7b4af32fa77
Author: Joao Eduardo Luis <joao@suse.de>
Date: Thu Jan 7 19:20:47 2016 +0000
mon: OSDMonitor: do not assume a session exists in send_incremental()
We may not have an open session for a given osd. If we blindly assume we
do, we may end up trying to send incrementals we do not have to the osd.
And then we will crash.
This fixes a regression introduced by
171fee1b82d2675e364da7f96dfb9dd286d9b6e6
which is meant as a backport of
de43a02e06650a552f048dc8acd17f255126fed9
but so happens to intruduce a line that wasn't on the original patch. We
imagine it was meant to make the 's->osd_epoch' assignment work without
checking the session, as per the original patch, but the backporter must
have forgotten to also backport the assertion on the not-null session.
The unfortunate introduction of the check for a not-null session
triggered this regression.
The regression itself is due to enforcing that a session exists for the
osd we are sending the incrementals to. However, if we come via the
OSDMonitor::process_failures() path, that may very well not be the case,
as we are handling potentially-old MOSDFailure messages that may no
longer have an associated session. By enforcing the not-null session, we
don't check whether we have the requested versions (i.e., if
our_earliest_version <= requested_version), and thus we end up on the
path that assumes that we DO HAVE all the necessary versions -- when we
may not, thus finally asserting because we are reading blank
incremental versions.
Fixes: #14236
Signed-off-by: Joao Eduardo Luis <joao@suse.de>
commit 4d0b9a1bbbabe7b27279a7b6e0a45f5b0d920c66
Author: Brad Hubbard <bhubbard@redhat.com>
Date: Mon Dec 7 11:31:28 2015 +1000
log: Log.cc: Assign LOG_DEBUG priority to syslog calls
Fixes: #13993
Signed-off-by: Brad Hubbard <bhubbard@redhat.com>
(cherry picked from commit 8e93f3f45db681f82633ca695a7dc4e7bd030584)
commit 26e832e76de90e0a751868b044ea745a97a5af82
Author: Jason Dillaman <dillaman@redhat.com>
Date: Fri Dec 18 15:22:13 2015 -0500
librbd: fix merge-diff for >2GB diff-files
Fixes: #14063
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
commit 9739d4de49f8167866eda556b2f1581c068ec8a7
Merge: d064636 a9d3f07
Author: Sage Weil <sage@redhat.com>
Date: Thu Dec 17 10:16:29 2015 -0500
Merge pull request #6544 from liewegas/wip-smaller-object-info
osd: make encoded object_info_t smaller to fit inside the XFS inode
Reviewed-by: Samuel Just <sjust@redhat.com>
commit a9d3f078cb6cdaf7f8778ba1a07d333548a9d1a2
Author: Sage Weil <sage@redhat.com>
Date: Mon Dec 14 13:13:33 2015 -0500
osd/osd_types: skip encoding newer object_info_t fields if they are unused
This reduces the size of the encoded object_info_t in most cases,
enough to get us under the 255 byte limit for a single inline
xattr in XFS.
Signed-off-by: Sage Weil <sage@redhat.com>
commit 1548a3fda7dd63e28792140e7e8ad1ac9b706e49
Author: Sage Weil <sage@redhat.com>
Date: Mon Dec 14 13:00:27 2015 -0500
osd/ReplicatedPG: do not set local_mtime on non-tiered pool
If a pool isn't tiered, don't bother with setting local_mtime. The only
users are the tiering agent (which isn't needed if there is not tiering)
and scrub for deciding if an object should get its digest recorded (we can
use mtime instead).
Signed-off-by: Sage Weil <sage@redhat.com>
commit 98bdb09bb8d632b2df0b252b3d9676fd9511a1c8
Author: Sage Weil <sage@redhat.com>
Date: Mon Dec 14 12:59:29 2015 -0500
osd/PGBackend: use mtime for digest decision if local_mtime is empty
If we don't have a local_mtime value, use mtime instead, for the purposes
of deciding if we should record a digest after scrub.
Signed-off-by: Sage Weil <sage@redhat.com>
commit fe1c28dea4e5607a0c72eab1f046074616cd55a7
Author: Sage Weil <sage@redhat.com>
Date: Mon Dec 14 11:35:46 2015 -0500
osd/ReplicatedPG: do not set omap digest if there is no omap
We want to avoid encoding it if we can. And if the FLAG_OMAP is not set
we don't need to *also* store an empty crc.
Signed-off-by: Sage Weil <sage@redhat.com>
commit 6cf261cd2e80adbaf2898cb354c8159a57749e9d
Author: xiexingguo <258156334@qq.com>
Date: Mon Nov 2 21:46:11 2015 +0800
Objecter: remove redundant result-check of _calc_target in _map_session.
Result-code check is currently redundant since _calc_target never returns a negative value.
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
(cherry picked from commit 5a6117e667024f51e65847f73f7589467b6cb762)
commit 5d0b5c1389d8c0baddb40cc5ff8f1d9cc6a4d315
Author: xiexingguo <258156334@qq.com>
Date: Thu Oct 29 17:32:50 2015 +0800
Objecter: potential null pointer access when do pool_snap_list.
Objecter: potential null pointer access when do pool_snap_list. Shall check pool existence first.
Fixes: #13639
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
(cherry picked from commit 865541605b6c32f03e188ec33d079b44be42fa4a)
commit 66ff0aa0130973aa2bea3a1a4ec8f4e3e6c19da8
Author: qiankunzheng <zheng.qiankun@h3c.com>
Date: Thu Nov 5 07:29:49 2015 -0500
auth/cephx: large amounts of log are produced by osd
if the auth of osd is deleted when the osd is running, the osd will produce large amounts of log.
Fixes:#13610
Signed-off-by: Qiankun Zheng <zheng.qiankun@h3c.com>
(cherry picked from commit 102f0b19326836e3b0754b4d32da89eb2bc0b03c)
commit 61da13bbe271964c56116fb2bed51045290f951a
Author: Chengyuan Li <chengyli@ebay.com>
Date: Thu Nov 19 22:29:39 2015 -0700
mon/PGMonitor: MAX AVAIL is 0 if some OSDs' weight is 0
In get_rule_avail(), even p->second is 0, it's possible to be used
as divisor and quotient is infinity, then is converted to an integer
which is negative value.
So we should check p->second value before calculation.
It fixes BUG #13840.
Signed-off-by: Chengyuan Li <chengyli@ebay.com>
(cherry picked from commit 18713e60edd1fe16ab571f7c83e6de026db483ca)
commit 53e81aab442855b34ee6e922f181bae3bf33e292
Author: wangchaunhong <root@A22832429.(none)>
Date: Tue Oct 20 18:40:23 2015 +0800
init-ceph: fix systemd-run cant't start ceph daemon sometimes
Fixes: #13474
Signed-off-by: Chuanhong Wang <wang.chuanhong@zte.com.cn>
(cherry picked from commit 2f36909e1e08bac993e77d1781a777b386335669)
Conflicts:
src/init-ceph.in
different content of cmd variable
commit c2c6d02591519dfd15ddcb397ac440322a964deb
Author: Piotr Dałek <piotr.dalek@ts.fujitsu.com>
Date: Mon Jul 6 09:56:11 2015 +0200
tools: fix race condition in seq/rand bench
Under certain conditions (like bench seq/rand -b 1024 -t 128) it is
possible that aio_read reads data into destination buffers before or
during memcmp execution, resulting in "[..] is not correct!" errors
even if actual objects are perfectly fine.
Also, moved latencty calculation around, so it is no longer affeted
by memcmp.
Signed-off-by: Piotr Dałek <piotr.dalek@ts.fujitsu.com>
Conflicts:
src/common/obj_bencher.cc
commit a619b621b0a7c670eeaf163d9e2b742d13c9f517
Author: Piotr Dałek <piotr.dalek@ts.fujitsu.com>
Date: Wed May 20 12:41:22 2015 +0200
tools: add --no-verify option to rados bench
When doing seq and rand read benchmarks using rados bench, a quite large
portion of cpu time is consumed by doing object verification. This patch
adds an option to disable this verification when it's not needed, in turn
giving better cluster utilization. rados -p storage bench 600 rand scores
without --no-verification:
Total time run: 600.228901
Total reads made: 144982
Read size: 4194304
Bandwidth (MB/sec): 966
Average IOPS: 241
Stddev IOPS: 38
Max IOPS: 909522486
Min IOPS: 0
Average Latency: 0.0662
Max latency: 1.51
Min latency: 0.004
real 10m1.173s
user 5m41.162s
sys 11m42.961s
Same command, but with --no-verify:
Total time run: 600.161379
Total reads made: 174142
Read size: 4194304
Bandwidth (MB/sec): 1.16e+03
Average IOPS: 290
Stddev IOPS: 20
Max IOPS: 909522486
Min IOPS: 0
Average Latency: 0.0551
Max latency: 1.12
Min latency: 0.00343
real 10m1.172s
user 4m13.792s
sys 13m38.556s
Note the decreased latencies, increased bandwidth and more reads performed.
Signed-off-by: Piotr Dałek <piotr.dalek@ts.fujitsu.com>
(cherry picked from commit ca6abca63de813c83a6960f83624be8e1a86a1f8)
Conflicts:
src/common/obj_bencher.cc
src/common/obj_bencher.h
commit d06463604cb8daeda288e824e8812352c0d6a7d9
Merge: ec35347 609f256
Author: Loic Dachary <loic@dachary.org>
Date: Wed Dec 2 20:48:00 2015 +0100
Merge pull request #6527 from theanalyst/wip-12856-hammer
rgw: missing handling of encoding-type=url when listing keys in bucket
Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com>
commit ec35347d8fd7a0eb175710e6111a9196746e278c
Merge: 487dc34 1b06f03
Author: Loic Dachary <loic@dachary.org>
Date: Wed Dec 2 20:47:43 2015 +0100
Merge pull request #6491 from SUSE/wip-13716-hammer
rgw:swift use Civetweb ssl can not get right url
Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com>
commit 487dc3436a8fabd1b1a31563071c907bdf18f525
Merge: 18d3ba3 99b4d1d
Author: Loic Dachary <loic@dachary.org>
Date: Wed Dec 2 20:47:27 2015 +0100
Merge pull request #6351 from theanalyst/wip-13538-hammer
rgw: orphan tool should be careful about removing head objects
Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com>
commit 18d3ba30323a44c1cc45bf4e870b8d9aeb5c3b0b
Merge: 36e4db3 6d89f4b
Author: Loic Dachary <loic@dachary.org>
Date: Wed Dec 2 20:47:09 2015 +0100
Merge pull request #6349 from theanalyst/wip-13540-hammer
rgw: get bucket location returns region name, not region api name
Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com>
commit 36e4db36f5ec9418672bbfe0dce6ae757f697efe
Merge: 8c4145e db1cbe7
Author: Loic Dachary <loic@dachary.org>
Date: Wed Dec 2 20:46:49 2015 +0100
Merge pull request #5910 from kmroz/wip-hammer-backport-content-type
rgw: backport content-type casing
Reviewed-by: Yehuda Sadeh <ysadehwe@redhat.com>
commit ac05617b246d3c2a329b1b99b0371e3f6b8541e9
Author: Sage Weil <sage@redhat.com>
Date: Thu Sep 17 21:42:53 2015 -0400
osd: fix send_failures() locking
It is unsafe to check failure_queue.empty() without the lock.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit b3ca828ae8ebc9068073494c46faf3e8e1443ada)
commit 8c4145ecc4a68accdb2120889fd933e8f6630dba
Merge: 4804eec 112c686
Author: Josh Durgin <jdurgin@redhat.com>
Date: Wed Nov 25 08:47:10 2015 -0800
Merge pull request #6587 from theanalyst/wip-13758-hammer
common: pure virtual method called
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
commit 4804eec095959c3747b13d29224b2cfa5b7d198d
Merge: 9b97e4f e693755
Author: Daniel Gryniewicz <dang@fprintf.net>
Date: Wed Nov 25 10:49:15 2015 -0500
Merge pull request #6352 from theanalyst/wip-13536-hammer
rgw: bucket listing hangs on versioned buckets
commit 9b97e4f6a41bb3fe7bae92b71ae266361022cf5c
Merge: 5a9e0a7 0378445
Author: Loic Dachary <loic@dachary.org>
Date: Wed Nov 25 16:48:09 2015 +0100
Merge pull request #6589 from theanalyst/wip-13693-hammer
osd: bug with cache/tiering and snapshot reads
Reviewed-by: Kefu Chai <kchai@redhat.com>
commit 5a9e0a722c654b83fc483e083ddb3035281f5397
Merge: 3047f2b a322317
Author: Loic Dachary <loic@dachary.org>
Date: Wed Nov 25 16:22:10 2015 +0100
Merge pull request #6585 from theanalyst/wip-13753-hammer
rbd: avoid re-writing old-format image header on resize
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
commit 3047f2bc50d5d26e70e832bd427f15073bf7733c
Merge: 407bd02 9c33dcc
Author: Loic Dachary <loic@dachary.org>
Date: Mon Nov 23 19:57:03 2015 +0100
Merge pull request #6586 from theanalyst/wip-13755-hammer
rbd: QEMU hangs after creating snapshot and stopping VM
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
commit 407bd0281582ebb3b50ce51bdb48244e22463c8f
Merge: c3c400f bddbda1
Author: Loic Dachary <loic@dachary.org>
Date: Mon Nov 23 17:32:21 2015 +0100
Merge pull request #6588 from theanalyst/wip-13770-hammer
Objecter: pool op callback may hang forever.
Reviewed-by: Sage Weil <sage@redhat.com>
commit c3c400f68d9c2593acd9485d1214d14af4e930d0
Merge: d116959 a52f7cb
Author: Loic Dachary <loic@dachary.org>
Date: Mon Nov 23 15:13:24 2015 +0100
Merge pull request #6430 from SUSE/wip-13654-hammer
crush: crash if we see CRUSH_ITEM_NONE in early rule step
Reviewed-by: Sage Weil <sage@redhat.com>
commit 3f1292e117ab1cdcaef3b24ee33854f7be142795
Author: Nathan Cutler <ncutler@suse.com>
Date: Thu Sep 3 20:30:50 2015 +0200
ceph.spec.in: fix License line
This is closer to my reading of
https://github.com/ceph/ceph/blob/master/COPYING than the previous version.
http://tracker.ceph.com/issues/12935 Fixes: #12935
Signed-off-by: Nathan Cutler <ncutler@suse.com>
(cherry picked from commit ef7418421b3748c712019c8aedd02b8005c1e1ea)
commit d116959442f67c8f36898ca989b490ca84a609c6
Merge: 12be099 74203b8
Author: Loic Dachary <loic@dachary.org>
Date: Mon Nov 23 09:28:09 2015 +0100
Merge pull request #6420 from SUSE/wip-13637-hammer
FileStore: potential memory leak if getattrs fails.
Reviewed-by: Kefu Chai <kchai@redhat.com>
commit 12be099a7b0f2692c167af96928e939b2227160c
Merge: 39c70d4 2052187
Author: Loic Dachary <loic@dachary.org>
Date: Sat Nov 21 09:10:27 2015 +0100
Merge pull request #6528 from theanalyst/wip-13695-hammer
init-rbdmap uses distro-specific functions
Reviewed-by: Boris Ranto <branto@redhat.com>
commit 39c70d4364a1be39f7c393847417f44279b4364c
Merge: 65aeba0 9643ee6
Author: Loic Dachary <loic@dachary.org>
Date: Sat Nov 21 09:07:13 2015 +0100
Merge pull request #6499 from SUSE/wip-13692-hammer
osd: do not cache unused memory in attrs
Reviewed-by: Samuel Just <sjust@redhat.com>
commit 65aeba0b2f165d893d06df7e9bbe25f989edafe9
Merge: 24d6698 394fbfc
Author: Loic Dachary <loic@dachary.org>
Date: Sat Nov 21 08:59:25 2015 +0100
Merge pull request #6143 from dachary/wip-13340-hammer
small probability sigabrt when setting rados_osd_op_timeout
Reviewed-by: Sage Weil <sage@redhat.com>
commit 24d66981d351848c44ae9e0395dad392f852be2b
Merge: 5d204db 171fee1
Author: Loic Dachary <loic@dachary.org>
Date: Sat Nov 21 08:58:11 2015 +0100
Merge pull request #5773 from tchaikov/wip-12835-hammer
mon: map_cache can become inaccurate if osd does not receive the osdmaps
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit a81bcf723c1099f2bea5daf8b01b7d9853de323a
Author: Jason Dillaman <dillaman@redhat.com>
Date: Thu Oct 15 00:15:54 2015 -0400
ceph_context: remove unsafe cast for singletons
It was previously assumed that a CephContext singleton would
inherit from CephContext::AssociatedSingletonObject, but it was
not enforced. This could result in unknown behavior when the
singleton is destroyed due to the implied virtual destructor.
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit fb62c78637d7092f48871d943282f45029bd6d29)
commit d50d7b2fa3751e8520694ee75eefe5ae56e57267
Author: Jason Dillaman <dillaman@redhat.com>
Date: Wed Sep 30 22:53:28 2015 -0400
osd: conditionally initialize the tracepoint provider
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 0009f343a5daf28358a669836d95660c0d9068d8)
commit bf34b3657339dc40c7939fcdddaf2b7ae78c82ad
Author: Jason Dillaman <dillaman@redhat.com>
Date: Wed Sep 30 22:37:05 2015 -0400
librados: conditionally initialize the tracepoint provider
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 6368c281b53d6175a564725b85516de4b6ae54de)
commit afc4f176f8371f0e61ce3463602f8a355b9283b9
Author: Jason Dillaman <dillaman@redhat.com>
Date: Wed Sep 30 22:10:10 2015 -0400
librbd: conditionally initialize the tracepoint provider
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 1a6eea95bd1ccef53f01cf53e493e652faa203ee)
Conflicts:
src/librbd/librbd.cc: trivial resolution
commit c82f93915875875b21ed779e7aaf297b8fbcdca1
Author: Jason Dillaman <dillaman@redhat.com>
Date: Wed Sep 30 19:17:08 2015 -0400
common: new tracing config options
Config options to enable LTTng-UST tracepoint providers for
the OSD, OSD objectstore, librados, and librbd.
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 2ccef89badb1f7a63721a0bf48d05eb4c2d0a97a)
Conflicts:
src/common/config_opts.h: trivial resolution
commit 2168c151066e0cea86284ffdf947a353f3b323d5
Author: Jason Dillaman <dillaman@redhat.com>
Date: Wed Sep 30 18:49:32 2015 -0400
tracing: merge tracepoint provider makefiles into main makefile
This results in the tracepoint provider shared libraries being
placed in the library path for unittests.
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit e11d8d68da27a1c224aa55fc40a72ac6aaa0441c)
Conflicts:
configure.ac: trivial resolution
src/Makefile.am: trivial resolution
commit d02beff1cf7650fe6e57cdafe64dcbee2631ed52
Author: Jason Dillaman <dillaman@redhat.com>
Date: Wed Sep 30 15:13:24 2015 -0400
tracing: dynamic tracepoint provider helper
The TracepointProvider class is a configuration observer. When
tracing is enabled, it will dynamically load the associated
tracepoint provider.
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit b3d02cc21b4cc40a194f84a9bbbf82cf9e7956d1)
Conflicts:
src/common/Makefile.am: trivial resolution
commit e53d66e42b1c3aea47832f7e8983284ec45d9efa
Author: Jason Dillaman <dillaman@redhat.com>
Date: Tue Sep 29 14:33:48 2015 -0400
packaging: add new tracepoint probe shared libraries
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit a7ed8e161b27c7852bf0f3a23a977ebd39137943)
commit bb7c0f501918b75dee97052faeafb1d6db61bd0a
Author: Jason Dillaman <dillaman@redhat.com>
Date: Tue Sep 29 14:30:43 2015 -0400
ceph.spec.in: add new tracepoint probe shared libraries
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit f4feee2f9dcd7b94476701b66d1f0bdf6fb6e0c2)
Conflicts:
ceph.spec.in: trivial resolution
commit e1da27134e4c9b4a2881aca664818598e5b2125b
Author: Jason Dillaman <dillaman@redhat.com>
Date: Tue Sep 29 14:13:46 2015 -0400
lttng: move tracepoint probes to dynamic libraries
LTTng-UST initializes itself at program load, which means it is
currently always enabled. This can lead to issues with SElinux
and AppArmor which might restrict access to the necessary device
files.
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 4a5305e41677885d45b2b74299113a3d00189c66)
Conflicts:
src/test/Makefile-client.am: trivial resolution
src/tracing/Makefile.am: trivial resolution
commit 5d204db85b2ed84411cfabea9eb73aa04c6ce05a
Merge: 1dbd4c6 8378aaf
Author: Loic Dachary <loic@dachary.org>
Date: Thu Nov 19 19:45:25 2015 +0100
Merge pull request #6580 from dachary/wip-13786-hammer
rbd-replay-* moved from ceph-test-dbg to ceph-common-dbg as well
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit 1dbd4c66307e615670c7569f107418876f8b977f
Merge: 24a5dc0 634d7f6
Author: Loic Dachary <loic@dachary.org>
Date: Thu Nov 19 19:45:03 2015 +0100
Merge pull request #6286 from dillaman/wip-13460-hammer
rbd-replay-prep and rbd-replay improvements
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit 24a5dc0a6b4f73c92c82e1933a1924db3f8e1090
Merge: d27da2e 6a40e4f
Author: Loic Dachary <loic@dachary.org>
Date: Thu Nov 19 19:41:20 2015 +0100
Merge pull request #5790 from SUSE/wip-12932-hammer
Miscellaneous spec file fixes
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit d27da2ee8fa3e074fdd91d414df2d59978db9497
Merge: 7d21127 8358fb8
Author: Kefu Chai <tchaikov@gmail.com>
Date: Thu Nov 19 19:13:11 2015 +0800
Merge pull request #6644 from dachary/wip-13812-upgrade
revert: osd: use GMT time for hitsets
Reviewed-by: Kefu Chai <kchai@redhat.com>
commit 8358fb8946d8809d695092baa4a6abf5d5b5e265
Author: Loic Dachary <loic@dachary.org>
Date: Wed Nov 18 18:08:58 2015 +0100
revert: osd: use GMT time for hitsets
"Merge pull request #5825 from tchaikov/wip-12848-hammer"
This reverts commit 39544718dc2f09bcfdc632ac72fd2a3cda87687e, reversing
changes made to 4ad97162026e1eb6e6e948ddf3eb39f711431e45.
http://tracker.ceph.com/issues/13812 Fixes: #13812
Signed-off-by: Loic Dachary <loic@dachary.org>
commit 4420929eacb67d17635fed53f15471a28942f31d
Author: brian <bjfelton@gmail.com>
Date: Wed Sep 23 09:49:36 2015 -0500
rgw: fix modification to index attrs when setting acls
Fixes: #12955
- add ACL change after initial population
- populate bufferlist with object attributes
Signed-off-by: Brian Felton <bjfelton@gmail.com>
(cherry picked from commit 7496741ebbd75b74d5ffeca5341cccb2318176e6)
commit 7d21127f433afa2d9172954e7b8ff47c40d2d62b
Merge: 6930601 1448915
Author: Loic Dachary <loic@dachary.org>
Date: Tue Nov 17 10:24:45 2015 +0100
Merge pull request #6402 from SUSE/wip-13621-hammer
CephFS restriction on removing cache tiers is overly strict
Reviewed-by: Greg Farnum <gfarnum@redhat.com>
commit 693060102594a2c505bed1e7e274731b554e9179
Merge: c2daf09 3f33ce6
Author: Loic Dachary <loic@dachary.org>
Date: Mon Nov 16 22:06:51 2015 +0100
Merge pull request #6354 from theanalyst/wip-13387-hammer
librbd: reads larger than cache size hang
Reviewed-by: Jason Dillaman <dillaman@redhat.com>
commit c2daf09fc6b8802240e13367a0a2f3349a7cca56
Merge: 1ca72fb af734e6
Author: Loic Dachary <loic@dachary.org>
Date: Mon Nov 16 19:48:07 2015 +0100
Merge pull request #6289 from dillaman/wip-13461-hammer
librbd: invalidate object map on error even w/o holding lock
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit 1ca72fbe6be0d664a3cbbd4356a8547c3b749a03
Merge: ca573fe 3e65730
Author: Loic Dachary <loic@dachary.org>
Date: Mon Nov 16 18:06:25 2015 +0100
Merge pull request #6153 from dachary/wip-13205-hammer
ReplicatedBackend: populate recovery_info.size for clone (bug symptom is size mismatch on replicated backend on a clone in scrub)
Reviewed-by: Samuel Just <sjust@redhat.com>
commit ca573fe7ac5a699b77cacb8fedfa50fffc77dd52
Merge: fdb3446 7161a2c
Author: Loic Dachary <loic@dachary.org>
Date: Mon Nov 16 18:06:12 2015 +0100
Merge pull request #6158 from dachary/wip-13336-hammer
osd: we do not ignore notify from down osds
Reviewed-by: Samuel Just <sjust@redhat.com>
commit fdb3446194743135b78a65c30a8285d001e563b6
Merge: f017566 0ad9521
Author: Loic Dachary <loic@dachary.org>
Date: Mon Nov 16 18:06:01 2015 +0100
Merge pull request #6335 from Abhishekvrshny/wip-13488-hammer
object_info_t::decode() has wrong version
Reviewed-by: Samuel Just <sjust@redhat.com>
commit f017566823af70323b7d4cbe1f7b3b761378d07e
Merge: a79acd4 9f3aebe
Author: Loic Dachary <loic@dachary.org>
Date: Mon Nov 16 18:05:39 2015 +0100
Merge pull request #6401 from SUSE/wip-13620-hammer
osd: pg stuck in replay
Reviewed-by: Samuel Just <sjust@redhat.com>
commit 8378aaf3f9d5463fb61a19d601201dd63a884419
Author: Loic Dachary <ldachary@redhat.com>
Date: Fri Nov 13 19:10:28 2015 +0100
build/ops: rbd-replay moved from ceph-test-dbg to ceph-common-dbg
http://tracker.ceph.com/issues/13785 Fixes: #13785
Signed-off-by: Loic Dachary <loic@dachary.org>
(cherry picked from commit a0204dc47acc0c50223e449fd23a2fc97bfd111a)
Conflicts:
debian/control: << 0.94.5-2 is appropriate for hammer
commit 03784457ab77617ddafc048b01044858b1e65bd3
Author: Kefu Chai <kchai@redhat.com>
Date: Tue Sep 29 22:26:48 2015 +0800
osd: fix the snapshot reads of evicted tiering pool
reset ssc->exsits in finish_ctx() if the ctx->cache_evict is true, and
the head is removed.
Fixes: #12748
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit b61f3e43f1c9d43daa0dedd7cbd0fe5787cbdfbb)
commit bddbda121c8d52ba8e50abd0341f14e332c6af5a
Author: xiexingguo <258156334@qq.com>
Date: Thu Oct 29 20:04:11 2015 +0800
Objecter: pool_op callback may hang forever.
pool_op callback may hang forever due to osdmap update during reply handling.
Fixes: #13642
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
(cherry picked from commit 00c6fa9e31975a935ed2bb33a099e2b4f02ad7f2)
commit 112c686ffc98e3739c9944635f4044e2b34f210a
Author: Jason Dillaman <dillaman@redhat.com>
Date: Tue Jul 7 12:11:13 2015 -0400
WorkQueue: new PointerWQ base class for ContextWQ
The existing work queues do not properly function if added to a running
thread pool. librbd uses a singleton thread pool which requires
dynamically adding/removing work queues as images are opened and closed.
Fixes: #13636
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 3e78b18b09d75626ca2599bac3b9f9c9889507a5)
Conflicts:
src/common/WorkQueue.h
Trivial merge conflict at class `ContextWQ` initialization
commit 9c33dccaad9a4cdd46e9ecfa1e3ba6c03d95885a
Author: Jason Dillaman <dillaman@redhat.com>
Date: Mon Nov 9 11:22:24 2015 -0500
librbd: fixed deadlock while attempting to flush AIO requests
In-flight AIO requests might force a flush if a snapshot was created
out-of-band. The flush completion was previously invoked asynchronously,
potentially via the same thread worker handling the AIO request. This
resulted in the flush operation deadlocking since it can't complete.
Fixes: #13726
Backport: infernalis, hammer
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit bfeb90e5fe24347648c72345881fd3d932243c98)
commit b3b7877f9b4b3f43acab09d0dd6ee971b6aa1c29
Author: Jason Dillaman <dillaman@redhat.com>
Date: Mon Nov 9 10:48:10 2015 -0500
tests: new test case to catch deadlock on RBD image refresh
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit a9729d9553e7fb925509cad8d388cf52a9fede9c)
commit a3223173f85c42147ff4ced730beffe85146a4ed
Author: Jason Dillaman <dillaman@redhat.com>
Date: Mon Nov 2 16:50:19 2015 -0500
librbd: resize should only update image size within header
Previously, the whole RBD image format 1 header struct was
re-written to disk on a resize operation.
Fixes: #13674
Backport: infernalis, hammer, firefly
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit d5be20b6d4646284571568ab28cbf45b0729390b)
commit 6a40e4f19d7b4cd45a25161303c7363e96fe799e
Author: Nathan Cutler <ncutler@suse.com>
Date: Tue Sep 29 10:39:16 2015 +0200
ceph.spec.in: lttng in SLES12 only
Over in the SUSE sector, we are trying to enable the SLE_12 and openSUSE_13.2
build targets. The lttng/babeltrace stuff is currently available only in
SLE_12.
Signed-off-by: Nathan Cutler <ncutler@suse.com>
(cherry picked from commit bf9ca1eb107c4462d8768faf3ff4db7972384ffd)
Conflicts:
ceph.spec.in
trivial resolution
commit e508a44485366557ac8a280de35584f3b5edf720
Author: Nathan Cutler <ncutler@suse.com>
Date: Wed Jul 15 12:54:20 2015 +0200
ceph.spec.in: fix lttng/babeltrace conditionals
lttng and babeltrace are build dependencies for rbd-replay-prep. Make
sure the right package names are used. Enable for SLE12, as well as
for openSUSE 13.1 and higher.
Move the BuildRequires out of the ceph-test subpackage and into the
distro-conditional dependencies section.
Make ordering of BuildRequires a little more alphabetical.
http://tracker.ceph.com/issues/12360 Fixes: #12360
Signed-off-by: Nathan Cutler <ncutler@suse.com>
(cherry picked from commit e267128ce22d8b8cd83f6d6d82f24f496600e678)
Conflicts:
ceph.spec.in
Dropped Requires: xmlstarlet in ceph-test that had inadvertently
been grabbed from d2cc2b1
commit 19c9546b79f506d0b4ee005a138a77f120c629e8
Author: Ken Dreyer <kdreyer@redhat.com>
Date: Tue Sep 8 10:41:02 2015 -0600
packaging: move rbd-replay* to ceph-common
The rbd-replay* utilities are useful for Ceph users with RBD clients.
Currently the rbd-replay* utilities ship in the "ceph-test" package, and
we intend this ceph-test package for Ceph developers and contributors,
not normal users.
Move the rbd-replay* utilities to "ceph-common".
http://tracker.ceph.com/issues/12994 Fixes: #12994
Signed-off-by: Ken Dreyer <kdreyer@redhat.com>
(cherry picked from commit 96486fd6f284ca3ab67f4f94631896d41c529e85)
Conflicts:
ceph.spec.in
50a33dea has not been backported to hammer
debian/ceph-test.install
50a33dea has not been backported to hammer
debian/control
different ceph-test Replaces: and Breaks: version
commit a79acd41187e6b049432bdc314f192e3fbb560a3
Merge: 3d61493 31b7864
Author: Loic Dachary <loic@dachary.org>
Date: Wed Nov 11 11:59:36 2015 +0100
Merge pull request #6213 from SUSE/wip-13425-hammer
wrong conditional for boolean function KeyServer::get_auth()
Reviewed-by: Kefu Chai <kchai@redhat.com>
commit 3d61493337d481dec5139ccb6172710a118b2009
Merge: 147f437 7ffd072
Author: Loic Dachary <loic@dachary.org>
Date: Wed Nov 11 06:59:55 2015 +0100
Merge pull request #6336 from Abhishekvrshny/wip-13535-hammer
LibRadosWatchNotify.WatchNotify2Timeout
Reviewed-by: Sage Weil <sage@redhat.com>
commit 147f437279e72c0766a126653e3db53abb73a4ac
Merge: 1970e61 9085c82
Author: Loic Dachary <loic@dachary.org>
Date: Wed Nov 11 06:59:42 2015 +0100
Merge pull request #6391 from SUSE/wip-13590-hammer
mon: should not set isvalid = true when cephx_verify_authorizer return false
Reviewed-by: Sage Weil <sage@redhat.com>
commit 1970e618209a242fc0bd3c707dfdfb1a886d6a2c
Merge: fb83704 ade0f1a
Author: Loic Dachary <loic@dachary.org>
Date: Wed Nov 11 06:58:55 2015 +0100
Merge pull request #6398 from SUSE/wip-13588-hammer
OSD::build_past_intervals_parallel() shall reset primary and up_primary when begin a new past_interval.
Reviewed-by: Sage Weil <sage@redhat.com>
commit 3d3595f86083a0f9847585eae4cffb8a82c816d4
Author: Jason Dillaman <dillaman@redhat.com>
Date: Tue Oct 27 10:13:27 2015 -0400
krbd: remove deprecated --quiet param from udevadm
This parameter has been removed since systemd 213, so this
effects Fedora 21+, Debian Jessie, and potentially future
releases of RHEL 7.
Fixes: #13560
Backport: hammer, infernalis
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 4300f2a9fe29627eea580564ff2d576de3647467)
commit 4d81cd19087e049bf3c2fe0d10de9215852ab51d
Author: Jason Dillaman <dillaman@redhat.com>
Date: Tue Oct 27 10:12:34 2015 -0400
run_cmd: close parent process console file descriptors
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit f46f7dc94139c0bafe10361622416d7dc343d31f)
commit fb83704d5db81aad95c61a61be52d07ce2d4d2b5
Merge: 3954471 0742177
Author: Loic Dachary <loic@dachary.org>
Date: Tue Nov 10 20:24:35 2015 +0100
Merge pull request #6199 from liewegas/wip-randomize-scrub-hammer
osd: randomize scrub times
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit 39544718dc2f09bcfdc632ac72fd2a3cda87687e
Merge: 4ad9716 6a4734a
Author: Loic Dachary <loic@dachary.org>
Date: Tue Nov 10 20:23:23 2015 +0100
Merge pull request #5825 from tchaikov/wip-12848-hammer
ReplicatedPG::hit_set_trim osd/ReplicatedPG.cc: 11006: FAILED assert(obc)
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit 2052187929e059a25f6a3baf67329f7ce0bf6d8a
Author: Boris Ranto <branto@redhat.com>
Date: Fri Oct 23 15:31:27 2015 +0200
init-rbdmap: Rewrite to use logger + clean-up
This patch rewrites the init-rbdmap init script so that it uses logger
instead of the log_* functions. The patch also fixes various smaller
bugs like:
* MAP_RV was undefined if mapping already existed
* UMNT_RV and UMAP_RV were almost always empty (if they succeeded) ->
removed them
* use of continue instead RET_OP in various places (RET_OP was not being
checked after the switch to logger messages)
* removed use of DESC (used only twice and only one occurrence actually
made sense)
Signed-off-by: Boris Ranto <branto@redhat.com>
(cherry picked from commit c567341e98fffbe39177f951511a7698f88abf5f)
commit f972f38445bc7a6efe122a5f9fc2ba7658625e26
Author: Boris Ranto <branto@redhat.com>
Date: Fri Oct 23 16:39:16 2015 +0200
ceph.spec.in: We no longer need redhat-lsb-core
Drop the redhat-lsb-core dependency as it is no longer necessary on
fedora/rhel.
The other two init scripts do not use redhat-lsb-core either. The
init-ceph.in conditionally requires /lib/lsb/init-functions and does not
use any of the functions defined in that file (at least not directly).
The init-radosgw file includes /etc/rc.d/init.d/functions on non-debian
platforms instead of /lib/lsb/init-functions file so it does not require
redhat-lsb-core either.
Signed-off-by: Boris Ranto <branto@redhat.com>
(cherry picked from commit 15600572265bed397fbd80bdd2b7d83a0e9bd918)
Conflicts:
ceph.spec.in
Merge conflict because there were other commits that introduced systemd
& SUSE specific bits in the master which havent yet reached
hammer. Since this commit only removed redhat-lsb-core from *el distros
removing the specific lines
commit 609f256a00f6ee18564ee055dd79f8293217940c
Author: Jeff Weber <jweber@cofront.net>
Date: Wed Aug 5 19:06:46 2015 -0400
rgw: implement s3 encoding-type for get bucket
This change introduces handling for the encoding-type request
parameter on the get bucket operation. An object key may contain
characters which are not supported in XML. Passing the value "url" for
the encoding-type parameter will cause the key to be urlencoded in the
response.
Fixes: #12735
Signed-off-by: Jeff Weber <jweber@cofront.net>
(cherry picked from commit 180ca7b56ba91a3995c76dac698aa4ee31d9a2ce)
commit 394fbfcc40d77ad6c8b3301c5ff2630c052107cd
Author: Ruifeng Yang <yangruifeng.09209@h3c.com>
Date: Fri Sep 25 10:18:11 2015 +0800
Objecter: maybe access wild pointer(op) in _op_submit_with_budget.
look at "after giving up session lock it can be freed at any time by response handler" in _op_submit,
so the _op_submit_with_budget::op maybe is wild after call _op_submit.
Fixes: #13208
Signed-off-by: Ruifeng Yang <yangruifeng.09209@h3c.com>
(cherry picked from commit 0635b1358354b19ae44105576f730381f3b5b963)
commit 84068f8cea0452333f00f8b65230c00caa2e52ff
Author: Ruifeng Yang <yangruifeng.09209@h3c.com>
Date: Fri Sep 25 12:42:28 2015 +0800
Objecter: repeated free op->ontimeout.
repeated free op->ontimeout in SafeTimer::timer_thread::callback->complete
Fixes: #13208
Signed-off-by: Ruifeng Yang <yangruifeng.09209@h3c.com>
(cherry picked from commit f1d8a8f577cee6d66f4dcffac667675f18145ebb)
commit 3e657304dc41facd40b8cab7531180083d14d22a
Author: Samuel Just <sjust@redhat.com>
Date: Fri Aug 28 12:46:57 2015 -0700
ReplicatedBackend::prepare_pull: set recover_info.size for clones
Fixes: #12828
Signed-off-by: Samuel Just <sjust@redhat.com>
(cherry picked from commit 1d3e46bedb9a8c0d2200e39f62f4e2c3337619f3)
commit 4ad97162026e1eb6e6e948ddf3eb39f711431e45
Merge: 92b2153 f1271ea
Author: Loic Dachary <loic@dachary.org>
Date: Tue Nov 10 10:34:01 2015 +0100
Merge pull request #6157 from dachary/wip-13040-hammer
common/Thread:pthread_attr_destroy(thread_attr) when done with it
Reviewed-by: Sage Weil <sage@redhat.com>
commit 92b21532f64792da023188d9aa620d900c0245ca
Merge: 1f02886 a17f2a9
Author: Loic Dachary <loic@dachary.org>
Date: Tue Nov 10 10:33:41 2015 +0100
Merge pull request #6155 from dachary/wip-13171-hammer
objecter: cancellation bugs
Reviewed-by: Sage Weil <sage@redhat.com>
commit 1f02886d7edc10ff93074afc3d2bf4cf77c709d5
Merge: 77df385 7d4b303
Author: Loic Dachary <loic@dachary.org>
Date: Tue Nov 10 10:33:19 2015 +0100
Merge pull request #6152 from dachary/wip-13233-hammer
mon: include min_last_epoch_clean as part of PGMap::print_summary and PGMap::dump
Reviewed-by: Sage Weil <sage@redhat.com>
commit 77df385e23452f401f3c0c269e220446d288fb8e
Merge: 469398f d8ca88d
Author: Loic Dachary <loic@dachary.org>
Date: Tue Nov 10 10:33:01 2015 +0100
Merge pull request #6146 from dachary/wip-13337-hammer
segfault in agent_work
Reviewed-by: Sage Weil <sage@redhat.com>
commit 469398f14b62af01068e65bcc5cc7c585a04e878
Merge: d87d136 397042a
Author: Loic Dachary <loic@dachary.org>
Date: Tue Nov 10 10:32:53 2015 +0100
Merge pull request #6145 from dachary/wip-13338-hammer
filestore: fix peek_queue for OpSequencer
Reviewed-by: Sage Weil <sage@redhat.com>
commit d87d1364fb8e345a1f85b65c2d4c7c80596ba5f9
Merge: 717eff9 d4e4d85
Author: Loic Dachary <loic@dachary.org>
Date: Tue Nov 10 10:32:42 2015 +0100
Merge pull request #6144 from dachary/wip-13339-hammer
mon: check for store writeablility before participating in election
Reviewed-by: Sage Weil <sage@redhat.com>
commit 717eff9a61818fde71ec5220998e9b5c14bbb3a9
Merge: e133ddc e8cce08
Author: Loic Dachary <loic@dachary.org>
Date: Tue Nov 10 10:30:54 2015 +0100
Merge pull request #5891 from Abhishekvrshny/wip-13035-hammer
requeue_scrub when kick_object_context_blocked
Reviewed-by: Sage Weil <sage@redhat.com>
commit e133ddc622d3532b0182c208f0ebb9956daab44f
Merge: c72306a 8ee93d5
Author: Loic Dachary <loic@dachary.org>
Date: Tue Nov 10 10:30:38 2015 +0100
Merge pull request #5890 from Abhishekvrshny/wip-13037-hammer
hit set clear repops fired in same epoch as map change -- segfault since they fall into the new interval even though the repops are cleared
Reviewed-by: Sage Weil <sage@redhat.com>
commit c72306a1dd3b193948d5df3371471797d4131cfe
Merge: 5e8c9d9 2bd5d0b
Author: Loic Dachary <loic@dachary.org>
Date: Tue Nov 10 10:21:07 2015 +0100
Merge pull request #6322 from sponce/WIP-13210-hammer
tests : BACKPORT #13210 Fixed broken Makefiles after integration of ttng into rados
Reviewed-by: Kefu Chai <kchai@redhat.com>
commit b62cac6c345df53bbda1c42a5336cffc8f0f9652
Author: Sangdi Xu <xu.sangdi@h3c.com>
Date: Wed Sep 23 19:31:32 2015 +0800
rgw: fix swift API returning incorrect account metadata
Fixes: #13140
Fix the bug that swift account stat command returns doubled object count and bytes used
Signed-off-by: Sangdi Xu <xu.sangdi@h3c.com>
(cherry picked from commit 66d19c78ba75b79190c81c95f94e7ef3084fda9e)
commit 9643ee65a6e968e479ca33a102f5f575924a7ff0
Author: Xinze Chi <xmdxcxz@gmail.com>
Date: Sun Aug 2 18:36:40 2015 +0800
bug fix: osd: do not cache unused buffer in attrs
attrs only reference the origin bufferlist (decode from MOSDPGPush or
ECSubReadReply message) whose size is much greater than attrs in recovery.
If obc cache it (get_obc maybe cache the attr), this causes the whole origin
bufferlist would not be free until obc is evicted from obc cache. So rebuild
the bufferlist before cache it.
Fixes: #12565
Signed-off-by: Ning Yao <zay11022@gmail.com>
Signed-off-by: Xinze Chi <xmdxcxz@gmail.com>
(cherry picked from commit c5895d3fad9da0ab7f05f134c49e22795d5c61f3)
commit 5e8c9d967e0ade393e36893965ca8ddfaa317b48
Merge: db0366d dba8b5b
Author: Loic Dachary <loic@dachary.org>
Date: Mon Nov 9 08:38:31 2015 +0100
Merge pull request #5810 from Abhishekvrshny/wip-12948-hammer
Heavy memory shuffling in rados bench
Reviewed-by: Kefu Chai <kchai@redhat.com>
commit 1b06f03b541983461469f8f53ef2a271b08d2f78
Author: Weijun Duan <duanweijun@h3c.com>
Date: Thu Oct 29 21:46:02 2015 -0400
rgw:swift use Civetweb ssl can not get right url
Fixes: #13628
Signed-off-by: Weijun Duan <duanweijun@h3c.com>
(cherry picked from commit e0fd540bf441e2f8276cbd96c601a0539892efe2)
commit db0366dd979cc0d401b1974c233e38dfe5b1b5d1
Merge: 22dce75 a65c398
Author: Loic Dachary <loic@dachary.org>
Date: Thu Nov 5 13:12:11 2015 +0100
Merge pull request #5530 from SUSE/wip-12587-hammer
FileStore calls syncfs(2) even it is not supported
Reviewed-by: Kefu Chai <kchai@redhat.com>
commit 22dce75181b254548226fc09decafac14dcf84a6
Merge: ed153c1 d8ac510
Author: Loic Dachary <loic@dachary.org>
Date: Wed Nov 4 19:38:53 2015 +0100
Merge pull request #5889 from Abhishekvrshny/wip-13042-hammer
ThreadPool add/remove work queue methods not thread safe
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
commit ed153c13226138eec71e472d1e3c8481db9b9009
Merge: ca8802f 8610de8
Author: Loic Dachary <loic@dachary.org>
Date: Wed Nov 4 19:14:28 2015 +0100
Merge pull request #6151 from dachary/wip-13245-hammer
client nonce collision due to unshared pid namespaces
Reviewed-by: Josh Durgin <jdurgin@redhat.com>
commit ca8802fb0cb8603267ecfc7320fee7a679617b8b
Merge: 19450b9 6e29e90
Author: Loic Dachary <loic@dachary.org>
Date: Wed Nov 4 11:36:39 2015 +0100
Merge pull request #6159 from dachary/wip-13341-hammer
ceph upstart script rbdmap.conf incorrectly processes parameters
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit 19450b90e4c20556580d276bab7dafb23db1758f
Merge: 18882c8 6849288
Author: Loic Dachary <loic@dachary.org>
Date: Wed Nov 4 11:34:07 2015 +0100
Merge pull request #6154 from dachary/wip-13195-hammer
should recalc the min_last_epoch_clean when decode PGMap
Reviewed-by: Sage Weil <sage@redhat.com>
commit 18882c81a71c4b517cb0eaa5a884637ea5cbe7a6
Merge: de4f37b 3c1f7cb
Author: Loic Dachary <loic@dachary.org>
Date: Wed Nov 4 11:28:03 2015 +0100
Merge pull request #6132 from SUSE/wip-13307-hammer
dumpling incrementals do not work properly on hammer and newer
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit de4f37bb0abc932c813983792e170de1bb40241b
Merge: 1c632a7 24268cf
Author: Loic Dachary <loic@dachary.org>
Date: Wed Nov 4 11:23:14 2015 +0100
Merge pull request #6097 from dillaman/wip-13045-hammer
librbd: diff_iterate needs to handle holes in parent images
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit 1c632a7ea7057522fb9df61994b5b3cf65416689
Merge: bedb334 71a42f7
Author: Loic Dachary <loic@dachary.org>
Date: Wed Nov 4 11:22:25 2015 +0100
Merge pull request #6060 from XinzeChi/wip-hammer-outdata-set
bug fix: osd: avoid multi set osd_op.outdata in tier pool
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit bedb3344f6c802ea04a171dabd4fb596284af34a
Merge: 0d3d819 cecd68d
Author: Loic Dachary <loic@dachary.org>
Date: Wed Nov 4 11:20:50 2015 +0100
Merge pull request #5897 from tchaikov/wip-12940-hammer
IO error on kvm/rbd with an erasure coded pool tier
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit 0d3d81951db61a97f6cb26d9f75e80a3400a2a37
Merge: 381cfcc 7475a8f
Author: Loic Dachary <loic@dachary.org>
Date: Wed Nov 4 11:11:00 2015 +0100
Merge pull request #5787 from SUSE/wip-12923-hammer
logrotate reload error on Ubuntu 14.04
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit 381cfccf38bade28a001f3f24392c61896b3c711
Merge: cad1cfb 86f88df
Author: Loic Dachary <loic@dachary.org>
Date: Wed Nov 4 11:10:26 2015 +0100
Merge pull request #5766 from dachary/wip-12590-hammer
ceph mds add_data_pool check for EC pool is wrong
Reviewed-by: Greg Farnum <gfarnum@redhat.com>
commit cad1cfb088b095e8333c4a7d79ccefc2b2c4d782
Merge: 9764da5 e8d6d5a
Author: Loic Dachary <loic@dachary.org>
Date: Wed Nov 4 11:06:18 2015 +0100
Merge pull request #5361 from ceph/wip-11786.hammer
mon: MonitorDBStore: get_next_key() only if prefix matches
Reviewed-by: Loic Dachary <ldachary@redhat.com>
commit 5105d50d1c20d465229189bbb24cdfb5d16cf7bc
Author: Loic Dachary <ldachary@redhat.com>
Date: Tue Nov 3 00:21:51 2015 +0100
tests: test/librados/test.cc must create profile
Now that the create_one_ec_pool function removes the testprofile each
time it is called, it must create the testprofile erasure code profile
again for the test to use.
http://tracker.ceph.com/issues/13664 Refs: #13664
Signed-off-by: Loic Dachary <ldachary@redhat.com>
(cherry picked from commit a60342942b5a42ee04d59af77a6b904ce62eefc4)
commit 302375068a59b86c50bda73ecefd99831ab52ea4
Author: Loic Dachary <ldachary@redhat.com>
Date: Mon Nov 2 20:24:51 2015 +0100
tests: destroy testprofile before creating one
The testprofile erasure code profile is destroyed before creating a new
one so that it does not fail when another testprofile erasure code
profile already exists with different parameters.
This must be done when creating erasure coded pools with the C++
interface, in the same way it's done with the C interface.
http://tracker.ceph.com/issues/13664 Fixes: #13664
Signed-off-by: Loic Dachary <ldachary@redhat.com>
(cherry picked from commit 47abab9a6f182aa0abe5047c04402850379bcd6d)
commit 3a7423dbe0a6af831d4c47acc98e6ced5ed5fe0c
Author: Loic Dachary <ldachary@redhat.com>
Date: Mon Nov 2 20:23:52 2015 +0100
tests: add destroy_ec_profile{,_pp} helpers
Signed-off-by: Loic Dachary <ldachary@redhat.com>
(cherry picked from commit ab46d79bc09fc711fa35302f49eecac81a98519b)
commit a52f7cb372339dffbeed7dae8ce2680586760754
Author: Sage Weil <sage@redhat.com>
Date: Tue Oct 13 09:55:01 2015 -0400
crush/mapper: ensure take bucket value is valid
Ensure that the take argument is a valid bucket ID before indexing the
buckets array.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 93ec538e8a667699876b72459b8ad78966d89c61)
commit 81d8aa14f3f2b7bf4bdd0b4e53e3a653a600ef38
Author: Sage Weil <sage@redhat.com>
Date: Tue Oct 27 20:55:26 2015 -0400
crush/mapper: ensure bucket id is valid before indexing buckets array
We were indexing the buckets array without verifying the index was within
the [0,max_buckets) range. This could happen because a multistep rule
does not have enough buckets and has CRUSH_ITEM_NONE
for an intermediate result, which would feed in CRUSH_ITEM_NONE and
make us crash.
Fixes: #13477
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 976a24a326da8931e689ee22fce35feab5b67b76)
commit 74203b806e8c268aa3edac8cfc7c4b393f7b1f27
Author: xiexingguo <258156334@qq.com>
Date: Mon Oct 26 18:38:01 2015 +0800
FileStore: potential memory leak if _fgetattrs fails
Memory leak happens if _fgetattrs encounters some error and simply returns.
Fixes: #13597
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
(cherry picked from commit ace7dd096b58a88e25ce16f011aed09269f2a2b4)
commit db1cbe700e41ee7a2db5ef10173c429283bc5a2a
Author: Robin H. Johnson <robin.johnson@dreamhost.com>
Date: Fri Sep 4 01:07:48 2015 +0000
Fix casing of Content-Type header
It turns out, despite the HTTP spec declaring that header field names
should be case-insensitive, some clients treat them wrongly, and
consider "Content-type" to not match "Content-Type".
CyberDuck was one of those clients, now fixed upstream in
https://trac.cyberduck.io/ticket/8999
To reduce future occurances of this bug, fix the casing of the
Content-Type header, to strictly comply with the HTTP specification (be
strict about what you send, and generous about what you receive).
Fixes: #12939
Backport: infernalis, hammer, firefly
Signed-off-by: Robin H. Johnson <robin.johnson@dreamhost.com>
(cherry picked from commit 1b9fbffdc24160251b96cec820d62fb2a12b6eab)
commit 1448915e49bb6c8abc192053e6f2e8a380c4e92a
Author: John Spray <john.spray@redhat.com>
Date: Mon Jun 1 13:55:22 2015 +0100
qa: update cephtool test for CephFS tier cases
1. Creating a filesystem using a
readonly tier on an EC pool (should be forbidden)
2. Removing a tier from a replicated base pool (should
be permitted)
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit f9b11b0abe7c2b0dcdfc5802336be8a0426ccb3a)
commit 14c36f0ccf103dea39b63e0e1b03082b99c42cb1
Author: John Spray <john.spray@redhat.com>
Date: Mon Jun 1 13:54:25 2015 +0100
mon: forbid readonly cache tiers for CephFS
For EC pools, we need a tier that is not just forwarding
writes, in order to provide the ops required
by CephFS.
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 11e5faf3c1d4fa28a8b41e3a92f0f684a844e5f5)
commit 31e59fbca0ba636fc57c208e612d7488a48a6287
Author: John Spray <john.spray@redhat.com>
Date: Thu May 7 14:23:37 2015 +0100
mon: refine check_remove_tier checks
Fixes: #11504
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit a50c8f1f2ad8845c7f77110868f9376f2d1ff883)
commit 9f3aebee16e256888b149fa770df845787b06b6e
Author: Sage Weil <sage@redhat.com>
Date: Tue Sep 22 13:57:37 2015 -0400
osd: fix requeue of replay requests during activating
If the replay period expires while we are still in the activating
state, we can simply insert our list of requests at the front of
the waiting_for_active list.
Fixes: #13116
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit d18cf51d9419819cdda3782b188b010969288911)
commit ade0f1a4285348c20fbae8e2a366fc332a7b7fbe
Author: xiexingguo <258156334@qq.com>
Date: Tue Oct 13 14:04:20 2015 +0800
OSD:shall reset primary and up_primary fields when beginning a new past_interval.
Shall reset primary and up_primary fields when we start over a new past_interval in OSD::build_past_intervals_parallel().
Fixes: #13471
Signed-off-by: xie.xingguo@zte.com.cn
(cherry picked from commit 65064ca05bc7f8b6ef424806d1fd14b87add62a4)
commit 9085c820491f25a04ec02accc5098c1ab9b57311
Author: yangruifeng <yangruifeng.09209@h3c.com>
Date: Mon Oct 19 08:08:12 2015 -0400
mon: should not set isvalid = true when cephx_verify_authorizer return false
Fixes: #13525
Signed-off-by: Ruifeng Yang <yangruifeng.09209@h3c.com>
(cherry picked from commit c7f75b8f7c0a773148ec16141941efd00ee76626)
commit 4875d05a3e265e80463a2a33a2c922affb0dad66
Author: Ken Dreyer <kdreyer@redhat.com>
Date: Mon Mar 16 10:02:35 2015 -0600
ceph.spec.in: rm EOL Fedoras; add OBS RHEL5 instead
Fedora 12 has been EOL for a long time. Remove the reference in the
RPM .spec file.
Since RHEL 5 support for Ceph is a work in progress, we won't remove
this entire python_sitelib / python_sitearch conditional for now, since
those are still needed on RHEL 5.
Add the rhel_version macro to make the conditional compatible with
SUSE's OBS.
Signed-off-by: Ken Dreyer <kdreyer@redhat.com>
(cherry picked from commit 353a3258968f76deaea4542d2165e9ab32b88de8)
commit dfba6acce75eea02251e98e2d57e6fbb344241fa
Author: Travis Rhoden <trhoden@redhat.com>
Date: Thu May 7 20:13:53 2015 -0400
packaging: Add rgw placeholder dirs in proper packages
Automatically create /var/lib/ceph/radosgw with the
ceph-radosgw package, and /var/lib/ceph/bootstrap-rgw
via the ceph package.
Signed-off-by: Travis Rhoden <trhoden@redhat.com>
(cherry picked from commit bf93128f664ac74383806530a1f974edde8df5b6)
Conflicts:
ceph.spec.in
hammer uses %ghost
commit 2387c8e0caef00e7181241e16549457d58a12192
Author: Nathan Cutler <ncutler@suse.cz>
Date: Fri May 15 21:39:39 2015 +0200
ceph.spec.in: consolidate centos/rhel macros
0%{?rhel} || 0%{?centos} can be simplified to just 0%{?rhel}, since
CentOS defines that macro for compatibility with RHEL.
Signed-off-by: Nathan Cutler <ncutler@suse.cz>
(cherry picked from commit 46404dd381e240634d224e77adc69f0ba4807045)
Conflicts:
ceph.spec.in
https://github.com/ceph/ceph/pull/4068 has not been backported
commit 7c58d312db2aa42088f419dbfa7e841d897fdad6
Author: Nathan Cutler <ncutler@suse.com>
Date: Thu Jun 25 16:55:10 2015 +0200
packaging: make logrotate a hard dependency across all distros
Given ceph's reputation for voluminous logs, it makes sense for
logrotate to be a hard dependency for all distros.
Signed-off-by: Nathan Cutler <ncutler@suse.com>
(cherry picked from commit 0734cd1ec3ca3bd012686dc1bdda675909224a49)
commit 61acc4e20ca58e11b9aba1efcdca78d1d7ddd2f7
Author: Nathan Cutler <ncutler@suse.com>
Date: Thu Jul 9 12:05:24 2015 +0200
ceph.spec.in: add missing -%{release}
We have it everywhere else and I can't think of any reason why
it should be omitted here.
Signed-off-by: Nathan Cutler <ncutler@suse.com>
(cherry picked from commit d952d59ae95d5f66aa4b98fd7aa4f4ecb36a4848)
commit 20590d367226ea01c9d4d95de15cfd1aa354d02b
Author: Nathan Cutler <ncutler@suse.com>
Date: Fri Jun 26 20:21:59 2015 +0200
ceph.spec.in: remove duplicate BuildRequires: sharutils
Signed-off-by: Nathan Cutler <ncutler@suse.com>
(cherry picked from commit c418bd9615e6f23f3c14ae9e71003c66d31d4b78)
commit 6b306eba829b1ec9822b24243ee7c021aab3d738
Author: Nathan Cutler <ncutler@suse.com>
Date: Mon Jun 15 12:58:00 2015 +0200
ceph.spec.in: fix python-flask dependency for SUSE
In SLE and openSUSE, the package is called python-Flask with an upper-case F.
Signed-off-by: Nathan Cutler <ncutler@suse.com>
(cherry picked from commit 39c28b9faf0fb069d91ac069a4d8df7bf03fb490)
Conflicts:
ceph.spec.in
add missing SUSE-specific build deps
commit 7a2cbcb0a4efa365b6d92ec58d7acf4a66e9d2b1
Author: Nathan Cutler <ncutler@suse.com>
Date: Thu Jun 25 18:55:39 2015 +0200
ceph.spec.in: clarify two important comments
First, the terms "common" and "specific" are vague. Second,
"common" can easily be confused with the ceph-common subpackage.
Fix this by rephrasing to "distro-unconditional dependencies" and
"distro-conditional dependencies", respectively.
Third, move the "distro-unconditional dependencies" header so it
is above the part where these dependencies are actually defined.
Signed-off-by: Nathan Cutler <ncutler@suse.com>
(cherry picked from commit dd212fd1d367f70c61f65a97599959668b44aa53)
commit dfe27c6ba05d49bb73854754b9d5c02cf1c7b1ef
Author: Ken Dreyer <kdreyer@redhat.com>
Date: Wed Jun 24 16:37:58 2015 -0600
ceph.spec.in: rm reference to EOL Fedoras (< 20)
Fedora 19 is EOL, so there's no need to refer to Fedoras less than 20.
"%{fedora}" effectively implies "Fedora 20 and above".
Reported-by: Ira Cooper <ira@samba.org>
Signed-off-by: Ken Dreyer <kdreyer@redhat.com>
(cherry picked from commit f3e4a91c458ca7b11706fe4608cb739499f5d3cc)
Conflicts:
ceph.spec.in
added xmlstarlet dependency to ceph-test package
commit 371b64ef4aa786db3fd29bc84ff98e63000a859b
Author: Ken Dreyer <kdreyer@redhat.com>
Date: Wed Jun 24 16:39:30 2015 -0600
ceph.spec.in: package rbd-replay-prep on all Fedoras
This reverts the change in commit
85517d611b7bf4cb6cbffcd2c65303be0d038264. Since we BuildRequire:
libbabeltrace-devel, autoconf will see that babeltrace is available
during the build, and make will build/install the rbd-replay-prep
utility.
This change also simplifies Fedora selection logic, because Fedora 19 is
EOL, so "%{fedora}" implies "Fedora 20 and above".
Reported-by: Ira Cooper <ira@samba.org>
Signed-off-by: Ken Dreyer <kdreyer@redhat.com>
(cherry picked from commit 2db9480fa2646de776f0076d82747e202e574ddd)
commit 82425b33f712047442d9d29c5a4349818b44b40a
Author: Boris Ranto <branto@redhat.com>
Date: Tue Jun 16 23:07:04 2015 +0200
ceph.spec.in: rbd-replay-prep is not being built on f21+
Signed-off-by: Boris Ranto <branto@redhat.com>
(cherry picked from commit 85517d611b7bf4cb6cbffcd2c65303be0d038264)
commit 53e2f702681dc08ddd0414b116e51de6488f42ca
Author: Owen Synge <osynge@suse.com>
Date: Mon Jun 8 17:48:55 2015 +0200
ceph.spec.in:BuildRequires sharutils
The uudecode binary is used to build Java-related components, and
uudecode is provided by the sharutils package on all supported
RPM platforms. When building with "--without=cephfs_java",
sharutils is not needed.
Thanks to Nathan Cutler <ncutler@suse.cz> for going into the
details with me.
On OBS without this patch we get the error message:
[ 170s] -----------------------------------------------------------------
[ 170s] ----- building ceph.spec (user abuild)
[ 170s] -----------------------------------------------------------------
[ 170s] -----------------------------------------------------------------
[ 170s] + exec rpmbuild -ba --define '_srcdefattr (-,root,root)' --nosignature --define 'disturl obs://build.opensuse.org/home:osynge:ceph:wip:wip_obs_fedora/Fedora_20/c0bbbc1e62228ca956ac3d367edc4fba-master' /home/abuild/rpmbuild/SOURCES/ceph.spec
[ 170s] error: Failed build dependencies:
[ 170s] sharutils is needed by ceph-1:2+git.1435043747.c1bd02c-1.1.x86_64
With this patch we can build fedora 22 and fedora 20 rpms fine.
Signed-off-by: Owen Synge <osynge@suse.com>
(cherry picked from commit 43c1784640538b813f0575c50f721ac911a30b3e)
Conflicts:
ceph.spec.in
_with_systemd -> trivial resolution
commit a84376c68499845de01f9e3b194b16a698125b6a
Author: Owen Synge <osynge@suse.com>
Date: Wed Jun 10 12:11:50 2015 +0200
ceph.spec.in: fix:Add missing directories breaking build
SUSE builds on OBS are failing with the missing dir entries:
/usr/share/ceph
/usr/lib/ceph
On suse these correspond to:
%dir %{_datarootdir}/ceph/
%dir %{_libexecdir}/ceph/
Signed-off-by: Owen Synge <osynge@suse.com>
(cherry picked from commit f1ff3e4dcb5dd7eddd2c346a05cfa78d860a5ce6)
commit 171fee1b82d2675e364da7f96dfb9dd286d9b6e6
Author: Kefu Chai <kchai@redhat.com>
Date: Tue Jul 21 00:33:19 2015 +0800
mon: track osd_epoch of all MonSessions
previously, we only track the osd_epoch for OSD peers. but other
MonClients who receives osdmaps can also benefit from osd_epoch.
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit de43a02e06650a552f048dc8acd17f255126fed9)
Conflicts:
src/mon/OSDMonitor.cc:
do not assume that the MonSession is always available when
OSDMonitor::send_incremental() is called. as when the
connection to the mon client is dropped, the Monitor is
aware of this, but not the OSDMonitor.
commit cc7da674965c3f758469e18788dc0e106e54d5bc
Author: Kefu Chai <kchai@redhat.com>
Date: Tue Jul 21 00:24:52 2015 +0800
mon: track osd_epoch in MonSession
* remove osd_epoch<osd, epoch> from OSDMonitor
* add osd_epoch to MonSession to track the latest osdmap epoch
OSDMonitor sends to a mon client
* do not remove osd_epoch entries if an OSD is down, or
max_osd > osd_id
Fixes: #10930
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit c05753eacc26e90b2e3b56e641a71bffd5b39bd0)
commit 3f33ce6132d5ce8b3454f30fbe14471ece39951d
Author: Lu Shi <shi.lu@h3c.com>
Date: Thu Sep 24 16:02:41 2015 +0800
librbd:reads larger than cache size hang.
Fixes:#13164
Signed-off-by: Lu Shi <shi.lu@h3c.com>
(cherry picked from commit 9c8200bb5d1ac9359803a182df03298b565b8479)
commit 5e5b512189b809bef503c34b0a19e30ed81cfbfa
Author: Radoslaw Zarzynski <rzarzynski@mirantis.com>
Date: Wed Aug 19 20:32:39 2015 +0200
rgw: url_decode values from X-Object-Manifest during GET on Swift DLO.
Fixes: #12728
Signed-off-by: Radoslaw Zarzynski <rzarzynski@mirantis.com>
(cherry picked from commit 69cf089596a0e5d4da2826b5c276306316a00690)
commit e693755a55e4da681a1947b1e1eb3bb537436cc0
Author: Yehuda Sadeh <yehuda@redhat.com>
Date: Tue Sep 22 14:53:42 2015 -0700
cls_rgw: fix bucket listing when dealing with invisible entries
Fixes: #12913
The bucket listing logic was broken when dealing with invisible entries.
Need to read keys until we can fetch all the requested entries, or
determine that there are no more entries. Anything else requires
a change to the protocol.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked from commit ef9730515e001d77637a77eb89950f70fd5e7f5a)
commit 99b4d1df3f6867e5ff0dd74691203737f6de6974
Author: Yehuda Sadeh <yehuda@redhat.com>
Date: Tue Sep 22 15:49:27 2015 -0700
rgw: orphan tool shouldn't clean up head objects
Fixes: #12958
Head objects are mutable, so removing them can race with object removal
and a later recreation, so we might end up cleaning them up when we don't
need to.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked from commit 030f697d196df9963ecf4f05f727da4798c583dc)
commit 6d89f4ba85063debe96734e9fdfcba247e770206
Author: Yehuda Sadeh <yehuda@redhat.com>
Date: Mon Oct 12 11:55:40 2015 -0700
rgw: location constraints should return api name
Fixes: #13458
The S3 location constraints request should return the api name,
not the region name.
Signed-off-by: Yehuda Sadeh <yehuda@redhat.com>
(cherry picked from commit 03078bab6ee8c617025902dbe52ec19e64653c33)
commit 2bd5d0b835b546c80b5af503d2b9538835c768fd
Author: Sebastien Ponce <sebastien.ponce@cern.ch>
Date: Tue Oct 20 09:22:16 2015 +0200
radosstriper : Fixed broken Makefiles after integration of lttng
Fixes: #13210, backport of #6322 into hammer
Signed-off-by: Sebastien Ponce <sebastien.ponce@cern.ch>
commit 7ffd072a8cc7e114d4d674d30333fea4db04dcdd
Author: Sage Weil <sage@redhat.com>
Date: Thu Oct 1 14:50:34 2015 -0400
osdc/Objecter: distinguish between multiple notify completions
We may send a notify to the cluster multiple times due to OSDMap
changes. In some cases, earlier notify attempts may complete with
an error, while later attempts succeed. We need to only pay
attention to the most-recently send notify's completion.
Do this by making note of the notify_id in the initial ACK (only
present when talking to newer OSDs). When we get a notify
completion, match it against our expected notify_id (if we have
one) or else discard it.
This is important because in some cases an early notify completion
may be an error while a later one succeeds.
Note that if we are talking to an old cluster we will simply not record a
notify_id and our behavior will be the same as before (we will trust any
notify completion we get).
Fixes: #13114
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit e86d033854c76f344c678e92016c4e5c5e0385e2)
Conflicts:
src/osdc/Objecter.cc
In Objecter::handle_watch_notify, a conflict was there due to a modified comment by commit 47277c51db7bb2725ea117e4e8834869ae93e006, which was not backported
commit 647c91098e8517912c73e3f7d27954e6f340472c
Author: Sage Weil <sage@redhat.com>
Date: Thu Oct 1 14:50:00 2015 -0400
osd: reply to notify request with our unique notify_id
The OSD assigns a unique ID to each notify it queues for
processing. Include this in the reply to the notifier so that
they can match it up with the eventual completions they receive.
This is necessary to distinguish between multiple completions
they may receive if there is PG peering and the notify is resent.
In particular, an earlier notify may return an error when a later
attempt succeeds.
This is forwards and backwards compatible: new clients will make use of
this reply payload but older clients ignore it.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 049ea702b9bf4a8a62ae2770d9ba55c0730b3eef)
commit 0ad9521944e0de5c135a066b0cda07f8976bdae1
Author: David Zafman <dzafman@redhat.com>
Date: Mon Oct 12 13:24:31 2015 -0700
osd: Correct the object_info_t::decode() version
Caused by: 13b9dc70
Fixes: #13462
Backport: Hammer, Firefly
Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 2b7ddde685213f3908f2dd9545b35f1d97f35a79)
commit af734e63f28ac5af414b6507c945ffd9e736d5cf
Author: Jason Dillaman <dillaman@redhat.com>
Date: Tue Oct 6 12:31:59 2015 -0400
librbd: invalidate object map on error even w/o holding lock
If there is a deep bug that prevents the object map from owning
the lock before objects are updated, flag the object map as invalid
regardless.
Fixes: #13372
Backport: hammer
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit e675400df7f8ba796d60e9bac4234857f0cb1392)
Conflicts:
src/librbd/ObjectMap.h/cc: hammer branch doesn't support rebuild
commit 634d7f62486847846e3862091b8cb587d095cb52
Author: Jason Dillaman <dillaman@redhat.com>
Date: Thu Oct 8 13:28:04 2015 -0400
tests: removed obsolete rbd_replay test cases
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit e0d8cb1283e4bdbfc770684b173e477c22558524)
commit 5c63d3c614189381cc169c64a3138c7abcb368c6
Author: Jason Dillaman <dillaman@redhat.com>
Date: Thu Oct 8 13:26:42 2015 -0400
ceph-dencoder: new rbd_replay trace file types
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit c2a83d0b518497dc8bcbbdea77b88febe52a3034)
commit a145a1cd8e245505abae66f0497be0df0078cecb
Author: Jason Dillaman <dillaman@redhat.com>
Date: Thu Oct 8 13:21:29 2015 -0400
rbd-replay: added version control to trace output file
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 3ecdae8388d69123b937a40ce614a0b795a757f1)
commit a30ed76c34ecca6278112908871126e1730a22bc
Author: Jason Dillaman <dillaman@redhat.com>
Date: Wed Oct 7 15:39:18 2015 -0400
rbd-replay-prep: added --verbose command line option
This new command-line will dump all the processed events as they
are extracted from the trace file.
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 646e50a771c3089121ec2f4369d1a8188001a3eb)
commit adf31c88d3ed95d0c29c649e2642ee3ae758743b
Author: Jason Dillaman <dillaman@redhat.com>
Date: Wed Oct 7 15:22:25 2015 -0400
rbd-replay-prep: stream events to the prep file
Avoid building the entire prep event history in memory before
attempting to write the prepped trace file.
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 98f513a113f4e8aff17c83991d2e6f0f3738a9c9)
commit 8c9e67ad0b21629565a09f6b9b92723b98594f84
Author: Jason Dillaman <dillaman@redhat.com>
Date: Wed Oct 7 14:56:22 2015 -0400
rbd-replay-prep: simplify IO dependency calculation
Only track read-after-write and write-after-write IO dependencies
via the associated write completions. All IO events after a write
completion are considered to be dependent and can be pruned down
to at most the number of concurrent IOs. This reduces the prep
time from a simple 'rbd bench-write' from over 4 hrs down to seconds.
Fixes: #13378, #13384
Backport: hammer
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 65fb1b86cbab65023f6207798c9e189bce55dcf6)
commit e7406fcba86646a4c3fbac66ee54b52ce3d5a33d
Author: Jason Dillaman <dillaman@redhat.com>
Date: Tue Oct 6 13:16:51 2015 -0400
rbd-replay: improve error messages
Fixes: #13221
Backport: hammer
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 4f436e10802d7b3c3949b592fec238c189a02c50)
commit 9f81d28107bd82b7118d1521ade982ab179c5a7a
Author: Jason Dillaman <dillaman@redhat.com>
Date: Tue Oct 6 12:48:50 2015 -0400
rbd-replay-prep: support new read/write APIs
Added support for read2, write2, aio_read2, and aio_write2 APIs.
Fixes: 13325
Backport: hammer
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit b5b4a9dd6bfe10844327a457849a0dbf7ef597ea)
commit a856bee2aa7e22f2bedf54b684c4d55452766b01
Author: Jason Dillaman <dillaman@redhat.com>
Date: Mon Oct 5 17:30:54 2015 -0400
rbd-replay-prep: avoid using assert statements on user inputs
Fixes: #13220
Backport: hammer
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit a1e99f06a78d7cf2ecc0c8e9fbc3331e4e4f18f7)
commit 6a4734a7afecb359b8bc9c114332cd7321097c38
Author: Samuel Just <sjust@redhat.com>
Date: Mon Sep 21 11:16:49 2015 -0700
ReplicatedPG::hit_set_setup: fix hit_set_remove_all call
We only want to do it if the pool config changed AND we are
primary && active.
fd38902dd4693b9b72f7532833e78f5df2c9efa5 partially fixed a related
bug. This should be backported along with it.
Fixes: 13192
Signed-off-by: Samuel Just <sjust@redhat.com>
(cherry picked from commit 36e4a80c39f7daab2a35efc04650ae8b7b11c4e0)
commit ab3877dd566f3951e76abaa6b7deeb5835a512b1
Author: Samuel Just <sjust@redhat.com>
Date: Tue Sep 22 11:13:44 2015 -0700
osd/: assert in HitSet constructor if type is TYPE_NONE as well
Otherwise impl is unpopulated and we risk segfaults in several
methods. It also seems like it would always indicate a bug.
Signed-off-by: Samuel Just <sjust@redhat.com>
(cherry picked from commit 8e5a801b43babf751b619ed9606d7be7e5b12837)
commit 4fe07b7f69321535a985958fc761a54cb71136f6
Author: Zhiqiang Wang <zhiqiang.wang@intel.com>
Date: Thu Jul 23 09:26:44 2015 +0800
osd: only remove all the hit set objects on primary osd
Only the primary PG is allowed to remove all the hit set objects. And
the PG should be in the active or peered states.
Signed-off-by: Zhiqiang Wang <zhiqiang.wang@intel.com>
(cherry picked from commit fd38902dd4693b9b72f7532833e78f5df2c9efa5)
commit 313a09b58b6aa6fe8999f4d5da07a04f00c64cd7
Author: Zhiqiang Wang <zhiqiang.wang@intel.com>
Date: Thu Jun 18 09:05:28 2015 +0800
osd: implement hit_set_remove_all
When hit set is not configured on startup or on a change, remove all
previous hitsets.
Signed-off-by: Zhiqiang Wang <zhiqiang.wang@intel.com>
(cherry picked from commit be28319bf3dc54b4b6f400d83405a29facfe3bd4)
Conflicts:
src/osd/ReplicatedPG.cc
pass the new argument of `p->using_gmt` to get_hit_set_archive_object()
commit b632254252f56d3b2c0c32803593ba765b45b222
Author: Kefu Chai <kchai@redhat.com>
Date: Tue Sep 8 16:52:32 2015 +0800
mon: disable gmt_hitset if not supported
the gmt_hitset is enabled by default in the ctor of pg_pool_t, this
is intentional. because we want to remove this setting and make
gmt_hitset=true as a default in future. but this forces us to
disable it explicitly when preparing a new pool if any OSD does
not support gmt hitset.
Fixes: #12968
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit e48cec3dc93b3988dcd8924933deb1b3a43e1d0f)
Conflicts:
src/mon/OSDMonitor.cc
remove unrelated hunk
commit 27fc9a3b2bd34bec8d5bb3954b873ce6344771f4
Author: Kefu Chai <kchai@redhat.com>
Date: Fri Aug 28 00:32:38 2015 +0800
osd: do not let OSD_HITSET_GMT reuse the feature bit
* to ease the backport to hammer
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 5a4f6a866bb675195ad83c38fd23238d078c78ed)
Conflicts:
src/include/ceph_features.h
minor changes to resolve conflicts
commit 7fcc61d26028eb417a0c224e8c9cd3cc95bc13d9
Author: David Zafman <dzafman@redhat.com>
Date: Thu Aug 27 11:24:25 2015 -0700
osd: Decode use_gmt_hitset with a unique version
Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 2bc5a48f4c5d3667213be3a7b5a0e0f5ef9daf4f)
Conflicts:
src/osd/osd_types.cc
minor changes to resolve conflicts
commit 26c7e968523ac09f608c4ed7c50681fbb741ca1f
Author: Kefu Chai <kchai@redhat.com>
Date: Tue Aug 18 16:53:50 2015 +0800
mon: print use_gmt_hitset in "ceph osd pool get"
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit cc2bcf760f2d2f20fc4a2fa78ba52475a23e55bf)
Conflicts:
src/mon/OSDMonitor.cc
use the old way to dump pool info
commit e8e00dab1b064e77a997504f385b5d60ee8aca25
Author: Kefu Chai <kchai@redhat.com>
Date: Tue Aug 18 00:04:23 2015 +0800
mon: add "ceph osd pool set $pool use_gmt_hitset true" cmd
allow "ceph osd pool set $pool use_gmt_hitset <true|1>" as long as
the cluster supports gmt hitset.
Fixes: #9732
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 03a1a3cf023a9aeb2fa26820e49e5efe3f3b3789)
commit 040e390d29fc68dcefe48c59cccacf6258c6f690
Author: Kefu Chai <kchai@redhat.com>
Date: Fri Jun 5 21:06:48 2015 +0800
osd: use GMT time for the object name of hitsets
* bump the encoding version of pg_hit_set_info_t to 2, so we can
tell if the corresponding hit_set is named using localtime or
GMT
* bump the encoding version of pg_pool_t to 20, so we can know
if a pool is using GMT to name the hit_set archive or not. and
we can tell if current cluster allows OSDs not support GMT
mode or not.
* add an option named `osd_pool_use_gmt_hitset`. if enabled,
the cluster will try to use GMT mode when creating a new pool
if all the the up OSDs support GMT mode. if any of the
pools in the cluster is using GMT mode, then only OSDs
supporting GMT mode are allowed to join the cluster.
Fixes: #9732
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 42f8c5daad16aa849a0b99871d50161673c0c370)
Conflicts:
src/include/ceph_features.h
src/osd/ReplicatedPG.cc
src/osd/osd_types.cc
src/osd/osd_types.h
fill pg_pool_t with default settings in master branch.
commit 31b7864b4161221f6027c6cbe1e42ed598d42b6a
Author: Nathan Cutler <ncutler@suse.com>
Date: Sun Sep 27 21:55:00 2015 +0200
CephxServiceHandler.cc: fix get_auth conditional
The KeyServer class has a public method get_auth() that returns a boolean
value. This value is being checked here - fix the conditional so it triggers
when get_auth() returns false.
http://tracker.ceph.com/issues/9756 References: #9756
Signed-off-by: Nathan Cutler <ncutler@suse.com>
(cherry picked from commit 7250db62cb7ac1e9c0b1f5956d60b9d874496702)
commit 0742177c767613d41bda0d260eebdd0087e3b392 (refs/remotes/me/wip-randomize-scrub-hammer)
Author: Kefu Chai <kchai@redhat.com>
Date: Sat May 16 00:07:27 2015 +0800
osd: use another name for randomize scrub option
s/osd_scrub_interval_limit/osd_scrub_interval_randomize_ratio/
Fixes: #10973
Signed-off-by: Kefu Chai <kchai@redhat.com>
commit fad33861a87f2feb944ad5454775c95d09364b56
Author: Kefu Chai <kchai@redhat.com>
Date: Mon Mar 9 16:42:34 2015 +0800
osd: randomize scrub times to avoid scrub wave
- to avoid the scrub wave when the osd_scrub_max_interval reaches in a
high-load OSD, the scrub time is randomized.
- extract scrub_load_below_threshold() out of scrub_should_schedule()
- schedule an automatic scrub job at a time which is uniformly distributed
over [now+osd_scrub_min_interval,
now+osd_scrub_min_interval*(1+osd_scrub_time_limit]. before
this change this sort of scrubs will be performed once the hard interval
is end or system load is below the threshold, but with this change, the
jobs will be performed as long as the load is low or the interval of
the scheduled scrubs is longer than conf.osd_scrub_max_interval. all
automatic jobs should be performed in the configured time period, otherwise
they are postponed.
- the requested scrub job will be scheduled right away, before this change
it is queued with the timestamp of `now` and postponed after
osd_scrub_min_interval.
Fixes: #10973
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 5e44040e8528bff06cc0a5a3f3293ab146e0e4e1)
Conflicts:
src/osd/OSD.cc
commit 9c130336a1a1dcf2338d202914aeacbc194af0b0
Author: Kefu Chai <kchai@redhat.com>
Date: Mon Mar 9 16:47:30 2015 +0800
osd: use __func__ in log messages
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 0f7f35670f03f3f58329c00d323963710e7e495d)
commit 4a3dedff8693953944af3eeb40994f38befebf48
Author: Kefu Chai <kchai@redhat.com>
Date: Wed Apr 15 12:04:05 2015 +0800
osd: simplify OSD::scrub_load_below_threshold() a little bit
avoid unnecessary comparison
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 2ab0e606dfd2847e1918be9f21f572d91befbf7b)
commit 24268cf4309267a0a1e32e28379cf7eafac00bca
Author: Jason Dillaman <dillaman@redhat.com>
Date: Fri May 1 14:21:35 2015 -0400
librbd: invoking RBD::open twice will leak memory
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit a5e88fcf11ebb16d4a1405b21cf6f895b37f185e)
commit 8610de81c8eeae763354a4133e160ccb3faa8dad
Author: Josh Durgin <jdurgin@redhat.com>
Date: Mon Sep 21 15:46:42 2015 -0700
use simplifed messenger constructor for clients
This is all mechanical except the calculation of the nonce, which is
now always randomized for clients.
Fixes: #13032
Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit a3a8c85b79afef67681c32c57b591c0e0a87a349)
commit 9d059c3f1b3aa0b1830b6f84694c850d36e8a77a
Author: Josh Durgin <jdurgin@redhat.com>
Date: Mon Sep 21 15:20:53 2015 -0700
msg: add simplified messenger constructor
Several callers create messengers using exactly the same parameters:
- reading the ms type from cct that is also passed in
- a default entity_name_t::CLIENT
- the default features
Additionally, the nonce should be randomized and not depend on
e.g. pid, as it does in several callers now. Clients running in
containers can easily have pid collisions, leading to hangs, so
randomize the nonce in this simplified constructor rather than
duplicating that logic in every caller.
Daemons have meaningful entity_name_ts, and monitors currently depend
on using 0 as a nonce, so make this simple constructer
client-specific.
Related: #13032
Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit e3785b0b5fbff870adbd5f121021b671912386cf)
Conflicts:
src/msg/Messenger.cc: Messenger::create only has 5 arguments in
hammer. Since the last one is zero, it does not change the
semantic.
commit 6e29e904ce8e7b3481fd3b318b070d326cd4a71a
Author: Sage Weil <sage@redhat.com>
Date: Wed Sep 30 08:29:05 2015 -0400
init-rbdmap: fix CMDPARAMS
Fixes: #13214
Reported-by: Wyllys Ingersoll <wyllys.ingersoll@keepertech.com>
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 02113ac449cd7631f1c9a3840c94bbf253c052bd)
commit 7161a2c927a6ded0d6ffe7d7621b7abd7f18acc3
Author: Samuel Just <sjust@redhat.com>
Date: Mon Sep 21 12:00:49 2015 -0700
PG: ignore info from down osd
Fixes: #12990
Backport: firefly, hammer
Signed-off-by: Samuel Just <sjust@redhat.com>
(cherry picked from commit 930d8eb1fb0daacd692d2e80b5ff473448bd4e8d)
commit 9753551f9f4ea9a1638dd469ef536d895ca9bc3b
Author: Samuel Just <sjust@redhat.com>
Date: Mon Sep 21 12:01:03 2015 -0700
OSDMap.h: add has_been_up_since
Signed-off-by: Samuel Just <sjust@redhat.com>
(cherry picked from commit 4b00350870e3a43aee5fff7982b3785fe1876634)
commit f1271eae0decf4d1fd9cfd83e5e54d3be6abfb17
Author: zqkkqz <zheng.qiankun@h3c.com>
Date: Fri Aug 7 10:49:45 2015 +0800
Common/Thread: pthread_attr_destroy(thread_attr) when done with it
When a thread attributes object is no longer required, it should be destroyed using the
pthread_attr_destroy() function. Destroying a thread attributes object has no effect on threads that were created using that object.
Fixes: #12570
Signed-off-by: zqkqkz zheng.qiankun@h3c.com
(cherry picked from commit 9471bb838a420ef5f439191d87e5388fecccb9e6)
commit 9581f9f4446ed5838bb3de6e1d7cf5faca85df90
Author: Piotr Dałek <piotr.dalek@ts.fujitsu.com>
Date: Fri Jul 17 12:43:52 2015 +0200
Thread.cc: remove malloc/free pair
There's no need for mallocing pthread_attr_t in Thread::try_create(),
it can be located on stack as it is freed in same function. This reduces
pressure put on memory manager.
Signed-off-by: Piotr Dałek <piotr.dalek@ts.fujitsu.com>
(cherry picked from commit 614ac0f4e375f3dd008c8937d3f89f6005d6e0a0)
commit a17f2a9a6600803a8b7f6dd306f7bafb6acc1af5
Author: Jianpeng Ma <jianpeng.ma@intel.com>
Date: Tue Sep 1 21:35:19 2015 +0800
osdc/Objecter: optimize Objecter::tick.
Set bool value is better than insert same value into set<>.
Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
(cherry picked from commit ee204044d9e69ea44b533c05cec154974039264c)
commit f80d237decac775d82f1e276a1ea600a9f352ffe
Author: Jianpeng Ma <jianpeng.ma@intel.com>
Date: Tue Sep 1 19:32:32 2015 +0800
osdc/Objecter: remove the unuseful code.
Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
(cherry picked from commit 80f10e3e59dadda2dca4eb62c68af972b701b316)
commit 8f76f4c467dc69cd2af75d9cbb134c26f2d04a84
Author: Jianpeng Ma <jianpeng.ma@intel.com>
Date: Tue Sep 1 15:39:29 2015 +0800
osdc/Objecter: Don't forget call _op_cancel_map_check when cancel linger op.
Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
(cherry picked from commit 7cc963b1c6ab37bf33638dc6eca7848d93f7908f)
commit ba36caba5226a44c41fdd2b19074916b38af5fdf
Author: Jianpeng Ma <jianpeng.ma@intel.com>
Date: Tue Sep 1 15:12:02 2015 +0800
osdc/Objecter: In _cancel_linger_op, it should make num_unacked/num_committed decrease.
Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
(cherry picked from commit 36b62710ddef0ffaee25837a92ca1ac9b353ff05)
commit 6ff7b02dc23b42a99c7136d1a55ead05c12e2b97
Author: Jianpeng Ma <jianpeng.ma@intel.com>
Date: Thu Aug 20 17:00:23 2015 +0800
osdc/Objeter: When cancel op, decrease num_unacked/num_uncommitted.
Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
(cherry picked from commit 064e8585a04edb3d87b38db6bed03e965cfcb359)
commit d815e0aacb4a516f4bbbb59a2b2073b38759d9e8
Author: Jianpeng Ma <jianpeng.ma@intel.com>
Date: Thu Aug 20 15:38:58 2015 +0800
osdc/Objecter: For func op_cancel_writes it can directly call op_cancel.
Becasue we get write-lock of rwlock, so it is safe to call op_cancel
rather than _op_canchel(homeless_session for this case don't met).
Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
(cherry picked from commit e4ce619fe17a7a9dfc18e6af0b84928aa2d88c00)
Conflicts:
src/osdc/Objecter.cc: the last if (found) was incorrectly
implemented as if (to_cancel.size()) after hammer. The
conflict is because the patch tries to revert an incorrect
if (to_cancel.size()) which is not in hammer.
commit 53316f3694c2a219541dcbdb0269ce97e2d2e9d7
Author: Jianpeng Ma <jianpeng.ma@intel.com>
Date: Wed Aug 19 14:54:21 2015 +0800
Objecter: Take RLocker when call is_active.
Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
(cherry picked from commit 89f0112e001a2561f9a5cd705898d43c8909501f)
commit 684928883f5eaaec99a8cc8c43e81fae69f9fbfa
Author: Kefu Chai <kchai@redhat.com>
Date: Wed Sep 16 15:08:17 2015 +0800
mon/PGMap: calc min_last_epoch_clean when decode
Fixes: #13112
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit d0ac68bf1785b330f3202d924c2203ace9393fe6)
commit 7d4b303b2d9649e2722a9b8f0be5c505776124df
Author: Guang Yang <yguang@yahoo-inc.com>
Date: Wed Sep 23 15:58:02 2015 +0000
osd: print min_last_epoch_clean along with pg dump
Fixes: 13198
Signed-off-by: Guang Yang <yguang@yahoo-inc.com>
(cherry picked from commit f5359f2314ce3e76ed446c01889dd23550c3ac36)
commit d8ca88dd4b47b4956211d206b0d42112e5842d3f
Author: Samuel Just <sjust@redhat.com>
Date: Tue Sep 22 11:44:18 2015 -0700
OSDService::agent_entry: don't use PG::operator<< without pg lock
Fixes: #13199
Signed-off-by: Samuel Just <sjust@redhat.com>
(cherry picked from commit d422f2820d5beee15117e166546a6386120e0349)
Conflicts:
src/osd/OSD.cc: trivial resolution
commit 397042a0cb2f4c628dd722192fee284359d2cbb3
Author: Xinze Chi <xinze@xksy.com>
Date: Wed Sep 23 19:26:00 2015 +0800
filestore: fix peek_queue for OpSequencer
Fixes: #13209
Signed-off-by: Xinze Chi <xinze@xsky.com>
(cherry picked from commit 5f7b3f5cde779f1300a35a04df2cdb1299f55472)
commit d4e4d85c005ff68cb0461e9d7f0cbe5e6cf2f073
Author: Sage Weil <sage@redhat.com>
Date: Wed Sep 23 10:58:01 2015 -0400
mon/Elector: do a trivial write on every election cycle
Currently we already do a small write when the *first* election in
a round happens (to update the election epoch). If the backend
happens to fail while we are already in the midst of elections,
however, we may continue to call elections without verifying we
are still writeable.
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit ef909ccbdc303cce8a39edef255325127832ff16)
commit 20a4c0c4831a9dcd49d8ad34b99369effbda9ab8
Author: Sage Weil <sage@redhat.com>
Date: Wed Sep 23 10:58:13 2015 -0400
mon/MonitorDBStore: assert/crash if there is a write error
Do this globally intead of relying on teh zillion mon callers to
check the error code. There are no cases where we want to
tolerate a commit failure.
Fixes: #13089
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 2fb7b1f0e33ada7c9a1be3de2f7708eb0760fcef)
commit 3c1f7cbc45c4744eea3c821f0343596756e05ab1
Author: Samuel Just <sjust@redhat.com>
Date: Thu Sep 24 18:35:39 2015 -0700
OSDMap: fill in known encode_features where possible
Otherwise, if we get an incremental from hammer (struct_v = 6) we will
encode the full map as if it were before CEPH_FEATURE_PGID64, which
was actually pre-argonaut. Similarly, if struct_v >= 7, we know it
was encoded with CEPH_FEATURE_OSDMAP_ENC.
Fixes: #13234
Backport: hammer
Signed-off-by: Samuel Just <sjust@redhat.com>
(cherry picked from commit 04679c5451e353c966f6ed00b33fa97be8072a79)
commit 93c523c65c1f0b6f9a76b1d71acf9e8df57c0fc9
Author: Jason Dillaman <dillaman@redhat.com>
Date: Tue Sep 8 15:47:37 2015 -0400
librbd: diff_iterate needs to handle holes in parent images
If a clone's parent image snapshot includes a discarded extent,
this was previously causing an assert failure. Instead, ignore
any discard holes in the parent image.
Fixes: #12885
Backport: hammer
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 3ccc3bb4bd35e57209852d460633e371b4d004e2)
Conflicts:
src/librbd/DiffIterate.cc : file doesn't exist in hammer, logic is in internal.cc
commit 359b68447393980bb7cdd4acc23582e8046b186c
Author: Jason Dillaman <dillaman@redhat.com>
Date: Mon Sep 28 10:25:07 2015 -0400
tests: new test case for librbd diff_iterate over discard extents
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit d5650c9cf85188efa73b279c8f4e4723fa475308)
Conflicts:
src/test/librbd/test_librbd.cc : diff_iterate2 doesn't exist in hammer
commit 71a42f7c3aa55de688ef16ea7e392573fa5a7e62
Author: Xinze Chi <xmdxcxz@gmail.com>
Date: Mon Jun 8 16:54:08 2015 +0800
bug fix: osd: avoid multi set osd_op.outdata in tier pool
There are two read op on the same object for ec pool. First op read
miss happend, calling do_proxy_read and promote_object, The second op only
do_proxy_read. but before first op process_copy_chunk finish, the second op
finish_proxy_read. first op receive reply from base pool
first and then second received. so the second op set the field "outdata"
in m->ops first. And then first op requeue_ops in process_copy_chunk,
At last in do_osd_ops, it append outdata field.
Fixes: 12540
Signed-off-by: Xinze Chi <xmdxcxz@gmail.com>
(cherry picked from commit 855ae1fd6e4557adba1cbd8ab532488b867cee2a)
Conflicts:
src/osd/ReplicatedPG.cc
commit e8cce089102abd8ec1b8e916c6f2c4a5b339cbd3
Author: Xinze Chi <xmdxcxz@gmail.com>
Date: Wed Jul 29 16:28:33 2015 +0800
bug fix: osd: requeue_scrub when kick_object_context_blocked
when read miss in writeback cache pool, osd do_proxy_read first
and maybe promote it. but in this case, the op is not added to
waiting_for_blocked_object. pg scrub maybe block by this object
(_range_available_for_scrub). so after promote it, we should
requeue_scrub.
Fixes: 12515
Signed-off-by: Xinze Chi <xmdxcxz@gmail.com>
(cherry picked from commit f77949fedce3449befd74efeb5270579f5085b16)
Conflicts:
src/osd/ReplicatedPG.cc
in ReplicatedPG::kick_object_context_blocked, removed hobject_t::BitwiseComparator from map p
commit 8ee93d5f661b303cafcb0134a0e9a1fce7b8e6b2
Author: Samuel Just <sjust@redhat.com>
Date: Thu Aug 27 11:08:33 2015 -0700
PG::handle_advance_map: on_pool_change after handling the map change
Otherwise, the is_active() checks in the hitset code can erroneously
return true firing off repops stamped with the new epoch which then get
cleared in the map change code. The filestore callbacks then pass the
interval check and call into a destroyed repop structure.
Fixes: 12809
Backport: hammer,firefly
Signed-off-by: Samuel Just <sjust@redhat.com>
(cherry picked from commit 14e02bc90a463805f4c3e2de210892067a52514b)
commit cecd68d06d4d114b0b735155da5c2bac736714c3
Author: Kefu Chai <kchai@redhat.com>
Date: Fri Aug 28 11:36:49 2015 +0800
ceph_test_rados_api_aio: add a test for aio_sparse_read
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 076bad955d374cbb37b77e2b0429f3c85f32abc0)
commit 8849f9933fe0fc315da0981f9cd88ea108f686ed
Author: Kefu Chai <kchai@redhat.com>
Date: Thu Aug 27 22:57:16 2015 +0800
ceph_test_rados_api_io: add tests for sparse_read
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 4d4920610ebfcb516630ed15678979c9e9292f5a)
Conflicts:
src/test/librados/test.cc
minor changes and remove C++11 stuff
commit 64b22dcba316bf9ac5598c249e4fba47c157f588
Author: Kefu Chai <kchai@redhat.com>
Date: Wed Aug 26 15:41:13 2015 +0800
ceph_test_rados: also send sparse_read in ReadOp
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 5ae2e7a185b5f95753a09a89d7110fc38848a083)
Conflicts:
src/test/osd/RadosModel.h
remove the piped-read test
commit 6ca593026f81580ecbd02ad6140347f31086a87d
Author: Kefu Chai <kchai@redhat.com>
Date: Fri Aug 28 14:27:53 2015 +0800
osd: should use ec_pool() when checking for an ecpool
we were using pool.info.require_rollback() in do_osd_ops() when
handling OP_SPARSE_READ to tell if a pool is an ecpool. should
use pool.info.ec_pool() instead.
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit a5bfde69a9d14de67da1e3354173ec70ba089b37)
commit 318049911445a3cc257db2b3beb3ab53e6f1d64e
Author: Kefu Chai <kchai@redhat.com>
Date: Thu Aug 27 22:57:49 2015 +0800
osd: translate sparse_read to read for ecpool
Fixes: #12012
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 700d42ef1c82f5602249b96690ae881c1d259d54)
commit d8ac5103e433f6649e28392b365d2e02b4ba1ffe
Author: Jason Dillaman <dillaman@redhat.com>
Date: Mon Aug 10 19:10:19 2015 -0400
WorkQueue: add/remove_work_queue methods now thread safe
These methods were not acquiring the ThreadPool lock when
manipulating the work_queue collection. This was causing
occasional crashes within librbd when opening and closing
images.
Fixes: #12662
Backport: hammer, firefly
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 3e18449b01c1ab78d1bbfc1cf111aa9bdbef7b1f)
commit e8d6d5a5f43b75f6aaec4c977221ad97d11e314c (refs/remotes/gh/wip-11786.hammer)
Author: Joao Eduardo Luis <joao@suse.de>
Date: Mon Sep 7 14:12:19 2015 +0100
mon: MonitorDBStore: make get_next_key() work properly
We introduced a significant bug with 2cc7aee, when we fixed issue #11786.
Although that patch would fix the problem described in #11786, we
managed to not increment the iterator upon returning the current key.
This would have the iterator iterating over the same key, forever and
ever.
Signed-off-by: Joao Eduardo Luis <joao@suse.de>
(cherry picked from commit 70d31082fd3dc8c7857994104577f1a3631c678c)
commit dba8b5b113f585586fcbbc0acf05543262486328
Author: Piotr Dałek <piotr.dalek@ts.fujitsu.com>
Date: Tue May 19 13:44:21 2015 +0200
tools: Don't delete, recreate and re-fill buffers in rados bench.
Fixes the high CPU usage and corrects rados bench scores on fast SSDs
and ramdisks/memstore.
For bench run on SSD, on Intel(R) Xeon(R) CPU E5-2640 v2 @ 2.00GHz
before this patch, times are:
write: real 5m0.169s, user 2m33.565s, sys 4m39.791s
seq: real 4m28.642s, user 1m35.250s, sys 6m42.948s
rand: real 5m0.258s, user 1m19.656s, sys 6m47.145s
After this patch:
write: real 5m1.162s, user 0m27.788s, sys 3m11.707s
seq: real 5m1.149s, user 2m23.278s, sys 4m14.427s
rand: real 5m1.021s, user 2m30.514s, sys 4m20.347s
Bench run: rados -p ssd bench 300 write|seq|read --no-cleanup
Note the increase in user time cpu on seq/read tests,
along with decreased sys cpu time; this is because there's
additional memcmp() that compares read objects with expected
contents. With less time spent memory juggling, more time is
spent performing more reads per second, increasing memcmp call
count and increasing amount of user cpu time used.
Signed-off-by: Piotr Dałek <piotr.dalek@ts.fujitsu.com>
(cherry picked from commit b894fc7acf7dee7f7ec8c5c280e7a6be41133328)
Conflicts:
src/common/obj_bencher.cc
return -5 was replaced with return r
commit fa72ecb509fa4ddd5f227b1e279eae390284a72c
Author: Piotr Dałek <piotr.dalek@ts.fujitsu.com>
Date: Fri May 15 15:44:18 2015 +0200
bufferlist: implement bufferlist::invalidate_crc()
This function iterates over all bufferlist internal buffers and calls
their invalidate_crc() method. Required for rados bench to work
correctly, because it modifies buffers outside buffer api, invalidating
internal CRC cache in the process - this method clears that cache, so
another call for buffer::get_crc() to return correct checksum.
Signed-off-by: Piotr Dałek <piotr.dalek@ts.fujitsu.com>
(cherry picked from commit 55a6f9efbee041d041742810ca4fa8874b6191a7)
commit 50c82f2095add5ad75200a62081fd2f5fea8d8d9
Author: Owen Synge <osynge@suse.com>
Date: Mon Jun 1 15:57:03 2015 +0200
ceph.spec.in summary-ended-with-dot
RPM spec files summary should not end with a dot.
This was inconsistent across the ceph rpm packages
and creates errors with rpm lint.
Signed-off-by: Owen Synge <osynge@suse.com>
(cherry picked from commit 042e8ff4dc71795d957bbfac12fade2bc0fc4674)
commit 94699e724b45a9a6de84a943c5b834d2f20f85ef
Author: Owen Synge <osynge@suse.com>
Date: Mon Jun 1 14:58:31 2015 +0200
ceph.spec.in libcephfs_jni1 has no %post and %postun
/usr/lib64/libcephfs_jni.so.1.0.0 requires /sbin/ldconfig to be
run after installing and after removing.
Signed-off-by: Owen Synge <osynge@suse.com>
(cherry picked from commit 0742d82b5561634a5f4ed73fc6cbe4d4f26b0d42)
commit 7475a8fe3f31a46a949f051ded383cd6f396a0fa
Author: Sage Weil <sage@redhat.com>
Date: Tue Apr 21 16:08:09 2015 -0700
logrotate.conf: prefer service over invoke-rc.d
On trusty 14.04, service works but invoke-rc.d does not (but is present).
Fixes: #11330
Reported-by: Wim Heirman <wim@heirman.net>
Signed-off-by: Sage Weil <sage@redhat.com>
(cherry picked from commit 6063a21f9edbbf80e44d45ff52a9f4ba1a1cb2c9)
commit 86f88dfce3c0e785d0ff48ec8892e53e6a53b080
Author: John Spray <john.spray@redhat.com>
Date: Thu Jul 23 10:15:56 2015 +0100
mon: fix checks on mds add_data_pool
Use same _check_pool helper as during fs creation.
Fixes: #12426
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 329741ec238d6af4b69bc838ba667be232741b3d)
commit a65c39852775b17323fa4efee54bde12ce22a512
Author: Kefu Chai <kchai@redhat.com>
Date: Wed Jul 29 15:52:19 2015 +0800
common/syncfs: fall back to sync(2) if syncfs(2) not available
Fixes: #12512
Signed-off-by: Kefu Chai <kchai@redhat.com>
(cherry picked from commit 08210d6a85527b0759f6dc4411f742bc13a4939f)
Conflicts:
src/common/sync_filesystem.h
- subtle difference in #if/#else/#endif structure
commit e6eaa914d685f2a8f1f8cf3b01112a1b720e1358
Author: Joao Eduardo Luis <joao@suse.de>
Date: Wed May 27 22:28:49 2015 +0100
mon: MonitorDBStore: get_next_key() only if prefix matches
get_next_key() had a bug in which we would always return the first key
from the iterator, regardless of whether its prefix had been specified
to the iterator.
Fixes: #11786
Signed-off-by: Joao Eduardo Luis <joao@suse.de>
(cherry picked from commit 2cc7aee1abe40453093306c8fef2312b650dff5d)